ACI 343R-95

(Reapproved 2004)

Analysis and Design of Reinforced

Concrete Bridge Structures
Reported by Joint ACI-ASCE Committee 343

John H. Clark Om P. Dixit

Chair Vice Chair
Hossam M. Abdou Ibrahim A. Ghais Antoine E. Naaman
John H. Allen Amin Ghali Andrzej S. Nowak
Gerald H. Anderson Joseph D. Gliken John C. Payne
F. Arbabi C. Stewart Gloyd Paul N. Roschke
Craig A. Ballinger Nabil F. Grace M. Saiid Saiidi
James M. Barker Hidayat N. Grouni Bal K. Sanan
Ostap Bender C. Donald Hamilton Harold R. Sandberg
T. Ivan Campbell Allan C. Harwood John J. Schemmel
Jerry Cannon Angel E. Herrera A. C. Scordelis
Claudius A. Carnegie Thomas T. C. Hsu Himat T. Solanki
John L. Carrato Ti Huang Steven L. Stroh
Gurdial Chadha Ray W. James Sami W. Tabsh
W. Gene Corley Richard G. Janecek Herman Tachau
W. M. Davidge David Lanning James C. Tai
H. Everett Drugge Richard A. Lawrie Marius B. Weschsler
William H. Epp James R. Libby J. Jim Zhao
Noel J. Everard Clellon L. Loveall
Anthony L. Felder W. T. McCalla

These recommendations, reported by Joint ACI-ASCE Committee 343 on
Concrete Bridge Design, provide currently acceptable guidelines for the
analysis and design of reinforced, prestressed, and partially prestressed
concrete bridges based on the state of the art at the time of writing the report.
The provisions recommended herein apply to pedestrian bridges, highway
bridges, railroad bridges, airport taxiway bridges, and other special bridge
structures. Recommendations for Transit Guideways are given in ACI 358R.
The subjects covered in these recommendations are: common terms;
general considerations; materials; construction; loads and load combina-
tions; preliminary design; ultimate load analysis and strength design;
service load analysis and design; prestressed concrete; superstructure
systems and elements; substructure systems and elements; precast
concrete; and details of reinforcement.
The quality and testing of materials used in construction are covered by
reference to the appropriate AASHTO and ASTM standard specifications.
Welding of reinforcement is covered by reference to the appropriate AWS
standard.
Keywords: admixtures; aggregates; anchorage (structural); beam-column
frame; beams (supports); bridges (structures); cements; cold weather
construction; columns (supports); combined stress; composite construction
(concrete and steel); composite construction (concrete to concrete);
compressive strength; concrete construction; concretes; concrete slabs;
ACI Committee Reports, Guides, Standard Practices, and Commentaries
are intended for guidance in planning, designing, executing, and inspecting
construction. This document is intended for the use of individuals who
are competent to evaluate the significance and limitations of its
content and recommendations and who will accept responsibility for the
application of the material it contains. The American Concrete Institute
disclaims any and all responsibility for the stated principles. The Institute
shall not be liable for any loss or damage arising therefrom.
Reference to this document shall not be made in contract documents. If
items found in this document are desired by the Architect/Engineer to be a
part of the contract documents, they shall be restated in mandatory
language for incorporation by the Architect/Engineer.
construction joints; construction materials; continuity (structural); cover;
curing; deep beams; deflection; earthquake-resistant structures; flexural
strength; footings; formwork (construction); frames; hot weather
construction; inspection; lightweight concretes; loads (forces); mixing;
mixture proportioning; modulus of elasticity; moments; placing; precast
concrete; prestressed concrete; prestressing steels; quality control; rein-
forced concrete; reinforcing steels; serviceability; shear strength; spans;
specifications; splicing; strength; structural analysis; structural design;
T-beams; torsion; ultimate strength method; water; welded-wire fabric.
Note: In the text, measurements in metric (SI) units in
parentheses follow measurements in inch-pound units.
Where applicable for equations, equations for metric (SI)
units in parentheses follow equations in inch-pound units.
CONTENTS
Chapter 1—Definitions, notation, and
organizations, p. 343R-4
1.1—Introduction
1.2—Definitions
1.3—Notation
1.4—Referenced organizations
Chapter 2—Requirements for bridges, p. 343R-12
2.1—Introduction
ACI 343R-95 became effective March 1, 1995 and supersedes ACI 343R-88. For
the 1995 revision, Chapters 6 and 12 were rewritten.
Copyright © 1995, American Concrete Institute.
All rights reserved including rights of reproduction and use in any form or by any
means, including the making of copies by any photo process, or by any electronic or
mechanical device, printed or written or oral, or recording for sound or visual repro-
duction or for use in any knowledge or retrieval system or device, unless permission in
writing is obtained from the copyright proprietors.

343R-1
343R-2 MANUAL OF CONCRETE PRACTICE

2.2—Functional considerations Chapter 8—Service load analysis and design,

2.3—Esthetic considerations 343R-96
2.4—Economic considerations 8.1—Basic assumptions
2.5—Bridge types 8.2—Serviceability requirements
2.6—Construction and erection considerations 8.3—Fatigue of materials
2.7—Legal considerations 8.4—Distribution of reinforcement in flexural members
8.5—Control of deflections
Chapter 3—Materials, p. 343R-26 8.6—Permissible stresses for prestressed flexural members
3.1—Introduction 8.7—Service load design
3.2—Materials 8.8—Thermal effects
3.3—Properties
3.4—Standard specifications and practices Chapter 9—Prestressed concrete, p. 343R-102
9.1—Introduction
Chapter 4—Construction considerations, 9.2—General design consideration
p. 343R-37 9.3—Basic assumptions
4.1—Introduction 9.4—Flexure, shear
4.2—Restrictions 9.5—Permissible stresses
4.3—Goals 9.6—Prestress loss
4.4—Planning 9.7—Combined tension and bending
4.5—Site characteristics 9.8—Combined compression and bending
4.6—Environmental restrictions 9.9—Combination of prestressed and nonprestressed rein-
4.7—Maintenance of traffic forcement—Partial prestressing
4.8—Project needs 9.10—Composite structures
4.9—Design of details 9.11—Crack control
4.10—Selection of structure type 9.12—Repetitive loads
4.11—Construction problems 9.13—End regions and laminar cracking
4.12—Alternate designs 9.14—Continuity
4.13—Conclusions 9.15—Torsion
9.16—Cover and spacing of prestressing steel
Chapter 5—Loads and load combinations, 9.17—Unbonded tendons
p. 343R-51 9.18—Embedment of pretensioning strands
5.1—Introduction
9.19—Concrete
5.2—Dead loads
9.20—Joints and bearings for precast members
5.3—Construction, handling, and erection loads
9.21—Curved box girders
5.4—Deformation effects
5.5—Environmental loads
Chapter 10—Superstructure systems and
5.6—Pedestrian bridge live loads elements, p. 343R-113
5.7—Highway bridge live loads 10.1—Introduction
5.8—Railroad bridge live loads 10.2—Superstructure structural types
5.9—Rail transit bridge live loads 10.3—Methods of superstructure analysis
5.10—Airport runway bridge loads 10.4—Design of deck slabs
5.11—Pipeline and conveyor bridge loads 10.5—Distribution of loads to beams
5.12—Load combinations 10.6—Skew bridges

Chapter 6—Preliminary design, p. 343R-66 Chapter 11—Substructure systems and elements,

6.1—Introduction p. 343R-123
6.2—Factors to be considered 11.1—Introduction
6.3—High priority items 11.2—Bearings
6.4—Structure types 11.3—Foundations
6.5—Superstructure initial section proportioning 11.4—Hydraulic requirements
6.6—Abutments 11.5—Abutments
6.7—Piers and bents 11.6—Piers
6.8—Appurtenances and details 11.7—Pier protection
6.9—Finishes
Chapter 12—Precast concrete, p. 343R-142
Chapter 7—Strength design, p. 343R-79 12.1—Introduction
7.1—Introduction 12.2—Precast concrete superstructure elements
7.2—Considerations for analysis, design, and review 12.3—Segmental construction
7.3—Strength requirements 12.4—Precast concrete substructures
 BRIDGE ANALYSIS AND DESIGN
12.5—Design
12.6—Construction
Chapter 13—Details of reinforcement for design
and construction, p. 343R-149
13.1—General
13.2—Development and splices of reinforcement
13.3—Lateral reinforcement for compression members
13.4—Lateral reinforcement for flexural members
343R-3
13.5—Shrinkage and temperature reinforcement
13.6—Standard hooks and minimum bend diameters
13.7—Spacing of reinforcement
13.8—Concrete protection for reinforcement
13.9—Fabrication
13.10—Surface conditions of reinforcement
13.11—Placing reinforcement
13.12—Special details for columns
343R-4 MANUAL OF CONCRETE PRACTICE
Complex highway interchange in California with 15 bridge structures
CHAPTER 1—DEFINITIONS, NOTATION, AND ORGANIZATIONS
1.1—Introduction
This chapter provides currently accepted definitions, nota-
tion, and abbreviations particular to concrete bridge design
practice which have been used in the preparation of this
document.
Concrete bridge types commonly in use are described
separately in Chapter 2, Requirements for Bridges, in
Chapter 6, Preliminary Design, and in Chapter 11, Super-
structure Systems and Elements.
1.2—Definitions
For cement and concrete terminology already defined,
reference is made to ACI 116R. Terms not defined in ACI
116R or defined differently from ACI 116R are defined for
specific use in this document as follows:
aggregate, normal weight—aggregate with combined
dry, loose weight, varying from 110 lb. to 130 lb/ft3 (approx-
imately 1760 to 2080 kg/m3).
compressive strength of concrete (fc′ )—specified
compressive strength of concrete in pounds per square inch
(psi) or (MPa).
Wherever this quantity is under a radical sign, the square
root of the numerical value only is intended and the resultant
is in pounds per square inch (psi) or (MPa).
concrete, heavyweight—a concrete having heavyweight
aggregates and weighing after hardening over 160 lb/ft3
(approximately 2560 kg/m3).
concrete, shrinkage-compensating—an expansive
cement concrete in which expansion, if restrained, induces
compressive strains that are intended to approximately offset
tensile strains in the concrete induced by drying shrinkage.
concrete, structural lightweight—concrete containing
lightweight aggregate having unit weight ranging from 90 to
115 lb/ft3 (1440 to 1850 kg/m3). In this document, a light-
weight concrete without natural sand is termed “all-light-
weight concrete,” and lightweight concrete in which all fine
aggregate consists of normal weight sand is termed “sand-
lightweight concrete.”
design load—applicable loads and forces or their related
internal moments and forces used to proportion members.
For service load analysis and design, design load refers to
loads without load factors. For ultimate load analysis and
strength design, design load refers to loads multiplied by
appropriate load factors.
effective prestress—the stress remaining in concrete due
to prestressing after all losses have occurred, excluding the
effect of superimposed loads and weight of member.
load, dead—the dead weight supported by a member
(without load factors).
load, live—the live load specified by the applicable docu-
ment governing design (without load factors).
load, service—live and dead loads (without load factors).
plain reinforcement—reinforcement without surface
deformations, or one having deformations that do not conform
to the applicable requirements for deformed reinforcement.
pretensioning—a method of prestressing in which the
tendons are tensioned before the concrete is placed.
surface water—water carried by an aggregate except that
held by absorption within the aggregate particles themselves.
 BRIDGE ANALYSIS AND DESIGN
1.3—Notation
Preparation of notation is based on ACI 104R. Where the
same notation is used for more than one term, the uncom-
monly used terms are referred to the Chapter in which they
are used. The following notations are listed for specific use
in this report:
a = depth of equivalent rectangular stress block
a = constant used in estimating unit structure dead load
(Chapter 5)
a = compression flange thickness (Chapter 7)
ab = depth of equivalent rectangular stress block for bal-
anced conditions
ai = fraction of trucks with a specific gross weight
av = ratio of stiffness of shearhead arm to surrounding
composite slab section
A = effective tension area of concrete surrounding the
main tension reinforcing bars and having the same
centroid as that reinforcement, divided by the num-
ber of bars, or wires. When the main reinforcement
consists of several bar or wire sizes, the number of
bars or wires should be computed as the total steel
area divided by the area of the largest bar or wire
used
A = axial load deformations and rib shortening used in
connection with t-loads (Chapter 5)
Ab = area of an individual bar
Ac = area of core of spirally reinforced compression
member measured to the outside diameter of the
spiral
Ae = area of longitudinal bars required to resist torsion
Ae = effective tension area of concrete along side face of
member surrounding the crack control reinforce-
ment (Chapter 8)
Af = area of reinforcement required to resist moment de-
veloped by shear on a bracket or corbel
Ag = gross area of section
Ah = area of shear reinforcement parallel to flexural ten-
sion reinforcement
Al = total area of longitudinal reinforcement to resist
torsion
An = area of reinforcement in bracket or corbel resisting
tensile force Nuc
Aps = area of prestressed reinforcement in tension zone
As = area of tension reinforcement
As′ = area of compression reinforcement
Asa = area of bonded reinforcement in tension zone
Ase = area of stirrups transverse to potential bursting
crack and within a distance se
Asf = area of reinforcement to develop compressive
strength of overhanging flanges of I- and T-sections
Ash = total area of hoop and supplementary cross ties in
rectangular columns
Ast = total area of longitudinal reinforcement (in com-
pression members)
At = area of one leg of a closed stirrup resisting torsion
within a distance s
Av = area of shear reinforcement within a distance s, or
area of shear reinforcement perpendicular to flexur-
343R-5
al tension reinforcement within a distance s, for
deep flexural members
Avf = area of shear-friction reinforcement
Avh = area of shear reinforcement parallel to the flexural
tension reinforcement within a distance s2
Aw = area of an individual wire
A1 = loaded area, bearing directly on concrete
A2 = maximum area of the portion of the supporting sur-
face that is geometrically similar to, and concentric
with, the loaded area
b = width of compressive face of member
b = constant used in estimating unit structure dead load
(Chapter 5)
b = width or diameter of pier at level of ice action
(Chapter 5)
b = width of web (Chapter 6)
b = width of section under consideration (Chapter 7)
be = width of concrete section in plane of potential burst-
ing crack
bo = periphery of critical section for slabs and footings
bv = width of the cross section being investigated for hor-
izontal shear
bw = web width, or diameter of circular section
B = buoyancy
c = distance from extreme compressive fiber to neutral
axis
C = construction, handling, and erection loads (Chapter 5)
C = stiffness parameter used in connection with lateral
distribution of wheel loads to multibeam precast
concrete bridges (Chapter 10)
C = ultimate creep coefficient (Chapter 5)
Ca = indentation coefficient used in connection with ice
forces
Ce = exposure coefficient used in connection with wind
forces
Ci = coefficient for pier inclination from vertical
Cm = factor used in determining effect of bracing on col-
umns (Chapter 7)
Ct = factor relating shear and torsional stress properties
equal to bw times d divided by the summation of x2
times y
Ct = creep deformation with respect to time (Chapter 5)
Cu = ultimate creep deformation (Chapter 5)
Cu = ultimate creep coefficient
Cw = shape factor relating to configuration of structure
and magnitude of wind force on structure
CF = centifugal force
d = distance from extreme compressive fiber to centroid
of tension reinforcement
d = depth of section under consideration (Chapter 7)
d = depth of girder (Chapter 5)
d′ = distance from extreme compressive fiber to cen-
troid of compression reinforcement
db = nominal diameter of bar, wire, or prestressing
strand
dc = thickness of concrete cover measured from the
extreme tensile fiber to the center of the bar lo-
cated closest thereto
343R-6 MANUAL OF CONCRETE PRACTICE

dp = effective depth of prestressing steel (Chapter 7) fl = stress produced by ith loading (Chapter 5)
ds = effective depth for balanced strain conditions flc = loss in prestressing steel stress due to creep
(Chapter 7) fle = loss in prestressing steel stress due to elastic
du = effective depth used in connection with pre- shortening
stressed concrete members (Chapter 7) flf = loss in prestressing steel stress due to friction
D = dead load flp = total loss in prestressing steel stress
D = diameter of lead plug in square or circular isola- flr = loss in prestressing steel stress due to relaxation
tion bearing (Chapter 11) fls = loss in prestressing steel stress due to shrinkage
Df = depth of footing fmin = algebraic minimum stress level where tension is
DF = distribution factor used in connection with live positive and compression is negative
loads fpc = compressive stress in the concrete, after all pre-
DR = derailment force stress losses have occurred, at the centroid of the
DS = displacement of supports cross section resisting the applied loads or at the
e = base of Napierian logarithms junction of the web and flange when the centroid
e = span for simply supported bridge or distance be- lies in the flange. (In a composite member, fpc will
tween points of inflection under uniform load be the resultant compressive stress at the centroid
(Chapter 10) of the composite section, or at the junction of the
e = eccentricity of design load parallel to axis mea- web and flange when the centroid lies within the
sured from the centroid of the section (Chapter 7) flange, due to both prestress and to bending mo-
eb = Mb/Pb = eccentricity of the balanced condition- ments resisted by the precast member acting
load moment relationship alone)
en = clear span length of slab (Chapter 10) fpe = compressive stress in concrete due to prestress
e1 = length of short span of slab only, after all losses, at the extreme fiber of a sec-
e2 = length of long span of slab tion at which tensile stresses are caused by ap-
E = effective width of concrete slab resisting wheel plied loads
or other concentrated load (Chapter 10) fpo = steel stress at jacking end of post-tensioning ten-
E = earth pressure used in connection with loads don
(Chapter 5) fps = stress in prestressing steel at design loads
Ec = modulus of elasticity of concrete fpu = ultimate strength of prestressing steel
Eci = modulus of elasticity of concrete at transfer of fpy = specified yield strength of prestressing tendons
stress fr = modulus of rupture of concrete
Eps = modulus of elasticity of prestressing strand fs = tensile stress in reinforcement at service loads
Es = modulus of elasticity of steel fs′ = stress in compressive reinforcement
EI = flexural stiffness of compression members fsb′ = stress in compressive reinforcement at balanced
EQ = earthquake force conditions
f = natural frequency of vibration of structure fse = effective stress in prestressing steel, after losses
(Chapter 5) ft = extreme fiber tensile stress in concrete at service
fa = axial stress loads
fa = basic allowable stress (Chapter 5) fy = specified yield stress, or design yield stress of
fb = bending stress nonprestressed reinforcement
fb = average bearing stress in concrete on loaded area fya = design yield stress of steel of bearing plate
(Chapter 8) fyh = design yield stress of steel for hoops and supple-
fc = extreme fiber compressive stress in concrete at mentary cross ties in columns
service loads F = frictional force
fc′ = specified compressive strength of concrete F = horizontal ice force on pier (Chapter 5)
fcds = change in concrete stress at center of gravity of Fa = allowable compressive stress
prestressing steel due to all dead loads except the Fb = allowable bending stress
dead load acting at the time the prestressing force g = acceleration due to gravity, 32.2 ft/sec2 (9.81
is applied m/sec2)
fci′ = compressive strength of concrete at time of ini- GA = ratio of stiffness of column to stiffness of mem-
tial prestress bers at A end resisting column bending
fcir = concrete stress immediately after transfer at cen- GA = degree of fixity in the foundation (Chapter 11)
ter of gravity of prestressing steel GB = ratio of stiffness of column to stiffness of mem-
fcp = concrete bearing stress under anchor plate of bers at B end resisting column bending
post-tensioning tendon Gavg = average ratio of stiffness of column to stiffness
fct = average splitting tensile strength of lightweight of members resisting column bending
aggregate concrete Gmin = minimum ratio of stiffness of column to stiff-
ff = stress range ness of members resisting column bending
 BRIDGE ANALYSIS AND DESIGN 343R-7

h = overall thickness of member ldh = development length for deformed bars in ten-
h = slab thickness (Chapter 6) sion terminating in a standard hook
h = height of rolled on transverse deformation of de- lhb = basic development length of hooked bar
formed bar (Chapter 8) ln = clear span measured face-to-face of supports
h = height of fill (Chapter 5) ln = length of tendon (Chapter 3)
h = thickness of ice in contact with pier (Chapter 5) lu = unsupported length of compression member
h = asphalt wearing surface thickness (Chapter 5) L = live load
ha = thickness of bearing plate L = span length used in estimating unit structure
hc = core dimension of column in direction under dead load (Chapter 5)
consideration L = bridge length contributing to seismic forces
hf = compression flange thickness of I- and T-sec- (Chapter 5)
tions L = length of compression member used in comput-
ho = thickness of standard slab used in computing ing pier stiffness (Chapter 11)
shrinkage LF = longitudinal force from live load
h2 = thickness of bottom slab of box girder (Chapter M = number of individual loads in the load combina-
6) tion considered
H = average height of columns supporting bridge M = live load moment per unit width of concrete
deck deck slab (Chapter 10)
H = curvature coefficient (Chapter 9) Ma = maximum moment in member at stage for which
I = impact due to live load (Chapter 5) deflection is being computed
Mb = nominal moment strength of a section at simul-
I = impact coefficient
taneous assumed ultimate strain of concrete and
I = moment of inertia (Chapter 7)
yielding of tension reinforcement (balanced
ICE = ice pressure
conditions)
Icr = moment of inertia of cracked section with rein-
Mc = factored moment to be used for design of com-
forcement transformed to concrete
pression member
Ie = effective moment of inertia for computation of
Mcr = moment causing flexural cracking at sections
deflection (Chapter 8)
due to externally applied loads
Ig = moment of inertia of gross concrete section Mm = modified moment (Chapter 7)
about the centroidal axis, neglecting the rein- Mmax = maximum factored moment due to externally
forcement applied loads, dead load excluded
Is = moment of inertia of reinforcement about the Mn = nominal moment strength of section
centroidal axis of the member cross section Mnx = nominal moment strength of section about x-
k = effective length factor for compression member axis
(Chapters 7 and 11) Mny = nominal moment strength of section about y-
k = dimensionless coefficient for lateral distribution axis
of live load for T- and I-girder bridge (Chapter Mu = factored moment at section, Mu = φ Mn
10) Mux = factored moment at section about x-axis, Mux =
k = coefficient for different supports in determining φ Mnx
earthquake force (Chapter 5) Muy = factored moment at section about y-axis, Muy =
ke = dimensionless coefficient for lateral distribution φ Mny
of live load for spread box-beam bridges (Chap- Mx = applied design moment component about x-axis
ter 10) My = applied design moment component about y-axis
K = wobble friction coefficient of prestressing steel M1 = value of smaller factored end moment on com-
(Chapter 9) pression member calculated from a convention-
K = constant used in connection with stream flow al or elastic analysis, positive if member is bent
(Chapters 5 and 11) in single curvature, negative if bent in double
K = value used for beam type and deck material curvature
(Chapter 10) M2 = value of larger factored end moment on com-
K = pier stiffness (Chapter 11) pression member calculated by elastic analysis,
l = length always positive
la = additional embedment length at support or at n = modular ratio Es/Ec
point of inflection n = number of individual loads in the load combina-
la = distance from face of support to load for brack- tion considered (Chapter 5)
ets and corbels (Chapter 7) nb = number of girders (Chapter 10)
lbd = basic development length for deformed bar in ne = number of design traffic lanes (Chapter 10)
compression N = nosing and lurching force
ld = development length N = minimum support length (Chapter 5)
343R-8 MANUAL OF CONCRETE PRACTICE

NB = number of beams RH = mean annual relative humidity, percent (Chap-

NL = number of design traffic lanes ter 5)
Nu = design axial load normal to the cross section oc- s = shear or torsion reinforcement spacing in direc-
curring simultaneously with Vu, to be taken as tion parallel to longitudinal reinforcement
positive for compression, negative for tension, s = beam spacing (Chapter 6)
and to include the effects of tension due to se = spacing of bursting stirrups
shrinkage and creep s2 = shear or torsion reinforcement spacing in direc-
Nuc = factored tensile force applied at top of bracket or tion perpendicular to the longitudinal reinforce-
corbel acting simultaneously with Vu, taken as ment or spacing of horizontal reinforcement in
positive for tension wall
o = overhang of bridge deck beyond supporting sw = spacing of wires
member (Chapter 6) S = span length
o = effective ice strength (Chapter 5) S = average beam spacing for distribution of live
OL = overload loads (Chapter 10)
p = allowable bearing
S = shrinkage and other volume changes used in
p = minimum ratio of bonded reinforcement in ten-
connection with loads or forces to be consid-
sion zone to gross area of concrete section
ered in analysis and design (Chapter 5)
(Chapter 9)
Sh = vertical spacing of hoops (stirrups) with a max-
p = unit weight of air (Chapter 5)
imum of 4 in. (Chapter 11)
p = proportion of load carried by short span of two-
Sh = spacing of hoops and supplementary cross ties
way slab (Chapter 10)
P = load on one rear wheel of truck equal to 12,000 SF = stream flow pressure = KV2
lb (53.4 kN) for HS15 loading and 16,000 lb SN = snow load
(71.1 kN) for HS20 loading (Chapter 10) t = actual time in days used in connection with
P = load above ground (Chapter 11) shrinkage and creep (Chapter 5)
Pb = design axial load strength of a section at simul- t = age of concrete in days from loading (Chapter 5)
taneous assumed ultimate strain of concrete and t* = equivalent time in days used in connection with
yielding of tension reinforcement (balanced shrinkage (Chapter 5)
conditions) tw = thickness of web in rectangular box section
Pcr = critical buckling load ty = temperature at distance y above depth of tem-
Pn = nominal axial load at given eccentricity perature variation of webs
Pnx = nominal axial load at given eccentricity about x- ty′ = temperature reduction for asphalt concrete
axis T = temperature
Pny = nominal axial load at given eccentricity about y- T = maximum temperature at upper surface of con-
axis crete (Chapter 5)
Pnxy = nominal axial load strength with biaxial loading T = fundamental period of vibration of the structure
Po = nominal load strength at zero eccentricity (Chapter 5)
Po = at rest earth pressures (Chapter 5) T* = minimum temperature of top slab over closed
Ps = ratio of spiral reinforcement interior cells (Chapter 5)
Pu = moment, shear, or axial load from the ith loading Tc = nominal torsional moment strength provided by
(Chapter 5) concrete
Pu = factored axial load at given eccentricity, Pu = φ Tn = nominal torsional moment strength
Pn Ts = nominal torsional moment strength provided by
Pux = factored axial load strength corresponding to torsional reinforcement
Mux with bending considered about the x-axis
Tu = factored torsional moment at section
only
v = total applied design shear stress at section
Puy = factored axial load strength corresponding to
vc = permissible shear stress carried by concrete
Muy with bending considered about the y-axis
only vdh = design horizontal shear stress at any cross sec-
Puxy = factored axial load strength with biaxial loading tion
q = dynamic wind pressure vh = permissible horizontal shear stress
r = radius of gyration of the cross section of com- vu = factored shear stress at section
pression member V = total applied design shear force at section
r = base radius of rolled on transverse deformation V = horizontal earthquake force (Chapter 5)
of deformed bar (Chapter 8) V = velocity of water used in connection with stream
R = average annual ambient relative humidity, percent flow (Chapter 5)
Rn = characteristic strength (moment, shear, axial V = maximum probable wind velocity (Chapter 5)
load) Vc = nominal shear strength provided by concrete
 BRIDGE ANALYSIS AND DESIGN
Vci = nominal shear strength provided by concrete
when diagonal cracking results from combined
shear and moment
Vcw = nominal shear strength provided by concrete
when diagonal cracking results from excessive
principal tensile stress in web
Vi = factored shear force at section due to externally
applied loads occurring simultaneously with
Mmax
Vn = nominal shear strength provided by concrete and
shear reinforcement
Vnh = nominal horizontal shear strength provided by
concrete and shear reinforcement
Vp = vertical component of effective prestress force at
section considered
Vs = nominal shear strength provided by shear rein-
forcement
Vu = factored shear force at section
w = unit structure dead load
wc = unit weight of concrete
wc = roadway width between curbs (Chapters 10 and
11)
we = road slab width from edge of slab to midway be-
tween exterior beam and first interior beam
W = wind load used in connection with application of
wind loads to different types of bridges
W = total weight of structure (Chapter 5)
W = crack width (Chapter 11)
Wf = gross weight of fatigue design truck
Wi = gross weight of specific trucks used in determin-
ing fatigue design truck
Wh = wind load applied in horizontal plane
Wp = weight of pier and footing below ground
Ws = weight of soil directly above footing
Wv = wind load applied in vertical plane
WL = wind load applied on live load (Chapter 5)
WL = wind load on live load
x = shorter overall dimension of rectangular part of
cross section
x = tandem spacing used in connection with aircraft
loads (Chapter 5)
x = width of box girder (Chapter 6)
x1 = shorter center-to-center dimension of closed
rectangular stirrup
x1 = distance from load to point of support (Chapter
10)
x2 = distance from center of post to point under inves-
tigation (Chapter 10)
y = longer overall dimension of rectangular part of
cross section
y = dual spacing used in connection with aircraft
loads (Chapter 5)
y = height of box girder (Chapter 6)
y1 = longer center-to-center dimension of closed rect-
angular stirrup
yd = mean thickness of deck between webs
343R-9
yt = distance from the centroidal axis of cross section,
neglecting the reinforcement, to the extreme fi-
ber in tension
Yo = depth of temperature variation of webs
Ys = height of temperature variation in soffit slab
z = quantity limiting distribution of flexural rein-
forcement
z = height of top of superstructure above ground
(Chapter 5)
α = angle between inclined shear reinforcement and
longitudinal axis of member
α = angle of pier inclination from vertical (Chapters
5 and 11)
α = load factor used in connection with group load-
ings (Chapter 5)
α = total angular change of prestressing steel profile
(Chapter 9)
αv = total vertical angular change of prestressing steel
profile (Chapter 9)
αh = total horizontal angular change of prestressing
steel profile (Chapter 9)
αf = angle between shear friction reinforcement and
shear plane
αi = load factor for the ith loading (Chapter 5)
αt = factor used in connection with torsion reinforce-
ment
β = percent of basic allowable stress (Chapter 5)
βb = ratio of area of bars cut off to total area of bars at
section
βc = ratio of long side to short side of concentrated
load or reaction area
βd = ratio of maximum factored dead load moment to
maximum factored total load moment, always
positive
β1 = factor used to determine the stress block in ulti-
mate load analysis and design
γ = unit weight of soil
δb = moment magnification factor for braced frames
δs = moment magnification factor for frames not
braced against sidesway
λ = correction factor related to unit weight of con-
crete
μ = coefficient of friction
μ = curvature friction coefficient (Chapter 9)
μ = ductility factor (Chapter 11)
ξ = time-dependent factor for sustained loads (Chap-
ter 8)
(ξcr)t = time-dependent factor for estimating creep under
sustained loads (Chapter 5)
ξi = instantaneous strain at application of load (Chap-
ter 5)
(ξsh)t = shrinkage at time t (Chapter 5)
(ξsh)u = ultimate shrinkage (Chapter 5)
ρ = ratio of tension reinforcement = As/bd
ρ′ = ratio of compression reinforcement = As′/bd
ρb = reinforcement ratio producing balanced condi-
tion
ρmin = minimum tension reinforcement ratio = As /bd
343R-10 MANUAL OF CONCRETE PRACTICE

ρp = ratio of prestressed reinforcement = Aps /bd AWS

ρs = ratio of volume of spiral reinforcement to total American Welding Society
volume of core (out-to-out of spirals) of a spiral- 550 NW LeJeune Road
ly reinforced compression member PO Box 351040
ρv = (As + Ah)/bd Miami, FL 33135
ρw = reinforcement ratio = As /bwd
σ = moment magnification factor for compression BPR
members Bureau of Public Roads
σ = effective ice strength (Chapter 5) This agency has been succeeded by the Federal Highway
τf = factor used in connection with prestressed con- Administration
crete member design (Chapter 7)
φ = strength-reduction factor CEB
Comite European du Beton
φ = angle of internal friction (Chapter 5)
(European Concrete Committee)
EPFL, Case Postale 88
1.4—Referenced organizations
This report refers to many organizations which are respon- CH 1015 Lausanne
sible for developing standards and recommendations for Switzerland
concrete bridges. These organizations are commonly
referred to by acronyms. Following is a listing of these orga- CRSI
nizations, their acronyms, full titles, and mailing addresses: Concrete Reinforcing Steel Institute
933 N. Plum Grove Road
Schaumburg, IL 60195
AASHTO
American Association of State Highway and Transportation
CSA
Officials
Canadian Standards Association
444 N. Capital Street, NW, Suite 225
178 Rexdale Boulevard
Washington, DC 20001 Rexdale (Toronto), Ontario
Canada M9W 1R3
ACI
American Concrete Institute FAA
PO Box 9094 Federal Aviation Administration
Farmington Hills, MI 48333-9094 800 Independence Avenue, SW
Washington, DC 20591
ANSI
American National Standards Institute FHWA
1439 Broadway Federal Highway Administration
New York, NY 10018 400 Seventh Street, SW
Washington, DC 20590
AREA
American Railway Engineering Association GSA
50 F Street, NW General Services Administration
Washington, DC 20001 18 F Street
Washington, DC 20405
ARTBA
HRB
American Road and Transportation Builders Association
Highway Research Board
525 School Street, SW
This board has been succeeded by the Transportation Re-
Washington, DC 20024
search Board
ASCE PCA
American Society of Civil Engineers Portland Cement Association
345 E. 47th Street 5420 Old Orchard Road
New York, NY 10017 Skokie, IL 60077

ASTM PCI
ASTM International Precast/Prestressed Concrete Institute
100 Barr Harbor Dr. 209 W. Jackson Blvd.
West Conshohocken, PA 19428 Chicago, IL 60606-6938
 BRIDGE ANALYSIS AND DESIGN
PTI
Post-Tensioning Institute
301 W. Osborn, Suite 3500
Phoenix, AZ 85013
TRB
Transportation Research Board
National Research Council
2101 Constitution Avenue, NW
Washington, DC 20418
343R-11
Recommended references
The documents of the various standards-producing organi-
zations referred to in this report are listed below with their
serial designation. The documents listed were the latest
effort at the time this report was written. Since some of these
documents are revised frequently, generally in minor detail
only, the user of this report should check directly with the
sponsoring group if it is desired to refer to the latest revision.
American Concrete Institute
104R Preparation of Notation for Concrete
116R Cement and Concrete Terminology
343R-12 MANUAL OF CONCRETE PRACTICE
O’Hare Field elevated roadway (photo courtesy of Alfred Benesch and Company).
CHAPTER 2—REQUIREMENTS FOR BRIDGES
2.1—Introduction
2.1.1 General—Design of bridge structures should be in
accord with requirements established by the owner, adapted to
the geometric conditions of the site and in accord with the
structural provisions of the applicable codes and specifications.
The geometry of the superstructure is dictated by the spec-
ified route alignment and the required clearances above and
below the roadway. These requirements are in turn directly
related to the type of traffic to be carried on the bridge deck,
as well as that passing under the bridge and, when the site is
near an airport, low flying aircraft. Thus, geometric require-
ments, in general, will be dependent on whether the bridge is
to carry highway, railway, transit, or airplane traffic, and
whether it is to cross over a navigable body of water, a
highway, a railway, or a transit route. Drainage, lighting, and
snow removal requirements should also be considered in the
geometric design of the superstructure.
Once the overall geometry of the superstructure has been
established, it should be designed to meet structural require-
ments. These should always include considerations of
strength, serviceability, stability, fatigue, and durability.
Before the reinforcing, prestressing, and concrete dimension
requirements can be determined, an analysis should be
performed to determine the internal forces and moments, the
displacements, and the reactions due to the specified loadings
on the bridge. This may be done using an elastic analysis, an
empirical analysis, or a plastic model analysis as described
in ACI SP-24. Because of their complexity, many bridge
structures have been analyzed by using an empirical
approach. However, by coupling modern day analytical tech-
niques with the use of digital computers, an elastic analysis of
even the most complex structural systems can now be
accomplished. Model analyses may prove useful when math-
ematical modeling is of doubtful accuracy, and especially in
cases where a determination of inelastic and ultimate
strength behavior is important.
2.1.2 Alignment—The horizontal and vertical alignment
of a bridge should be governed by the geometrics of the
roadway or channels above and below.
If the roadway or railway being supported on the bridge is
on a curve, the most esthetic structure is generally one where
the longitudinal elements are also curved. Box girders and
slabs, if continuous, are readily designed and built on a
curve. Stringers and girders can be curved but are more diffi-
cult to design and construct. If the curve is not sharp, the
girders or stringers can be constructed in straight segments
with the deck constructed on a curve. In this case the
following points require close examination:
a. Nonsymmetrical deck cross section.
b. Deck finish of the “warped” surface.
c. Vertical alignment of curbs and railing to preclude
visible discontinuities.
d. Proper development of superelevation.
Arches, cable-stayed, and suspension bridges are not
easily adaptable to curved alignments.
 BRIDGE ANALYSIS AND DESIGN 343R-13

2.1.3 Drainage—The transverse drainage of the roadway

should be accomplished by providing a suitable crown or
superelevation in the roadway surface, and the longitudinal
drainage should be accomplished by camber or gradient.
Water flowing downgrade in a gutter section of approach
roadway should be intercepted and not permitted to run onto
the bridge. Short continuous span bridges, particularly over-
passes, may be built without drain inlets and the water from the
bridge surface carried off the bridge and downslope by open or
closed chutes near the end of the bridge structure. Special
attention should be given to insure that water coming off the
end of the bridge is directed away from the structure to avoid
eroding the approach embankments. Such erosion has been a
source of significant maintenance costs.
Longitudinal drainage on long bridges is accomplished by
providing a longitudinal slope of the gutter (minimum of 0.5
percent preferred) and draining to scuppers or inlets which
should be of a size and number to drain the gutters
adequately. The positions of the scuppers may be determined
by considering a spread of water of about one-half a lane
width into the travel lane as recommended in “Drainage of
Highway Pavements.”2-1 At a minimum, scuppers should be
located on the uphill side of each roadway joint. Down-
spouts, where required, should be of rigid corrosion-resistant
material not less than 4 in. (100 mm) and preferably 6 in.
(150 mm) in the least dimension and should be designed to
be easily cleaned. The details of deck drains and downspouts
should be such as to prevent the discharge of drainage water
against any portion of the structure and to prevent erosion at
the outlet of the downspout.
Overhanging portions of concrete decks should be
provided with a drip bead or notch within 6 in. (150 mm) of
the outside edge.
2.2—Functional considerations
2.2.1 Highway bridges
2.2.1.1 Highway classification—Highways are classi-
fied by types for their planning, design, and administration.
The classification in each jurisdiction is made in accordance
with the importance of the highway, the traffic volume, the
design speed, and other pertinent aspects. The following
functional considerations are dependent upon the highway
classifications:
2.2.1.2 Width—The roadway width (curb-to-curb, rail-
to-rail, or parapet-to-parapet distance) is dependent on the
number of traffic lanes, the median width, and the shoulder
width. The preferred roadway width should be at least that
distance between approach guardrails, where guardrails are
provided, or the out-to-out approach roadway, and shoulder
width as recommended in AASHTO HB-12. Reduced
widths are sometimes permitted where structure costs are
unusually high or traffic volumes unusually low. Where
curbed roadway sections approach a structure, the same
section should be carried across the structure.
Recommendations as to roadway widths for various
volumes of traffic are given in AASHTO DS-2, DSOF-3,
GD-2 and GU-2.
Fig. 2.2.1.3—Clearance diagram for bridges.
2.2.1.3 Clearances—The horizontal vehicular clearance
should be the clear width measured between curbs or sidewalks,
and the vertical clearance should be the clear height for the
passage of vehicular traffic measured above the roadway at the
crown or high point of superelevation (Fig. 2.2.1.3).
Unless otherwise provided, the several parts of the struc-
ture should be constructed to secure the following limiting
dimensions or clearances for traffic:
a. The minimum horizontal clearance for low traffic
speed and low traffic volume bridges should be 8 ft
(2.4 m) greater than the approach travelled way. The
clearance should be increased as speed, type, and
volume of traffic dictate in accordance with AASHTO
DS-2, DSOF-3, GD-2, and GU-2.
b. Vertical clearance on state trunk highways and inter-
state systems in rural areas should be at least 16 ft (5 m)
over the entire roadway width, to which an allowance
should be added for resurfacing. On state trunk high-
ways and interstate routes through urban areas, a 16-ft
(5-m) clearance should be provided except in highly
developed areas. A 16-ft (5-m) clearance should be
provided in both rural and urban areas, where such
clearance is not unreasonably costly and where needed
for defense requirements. Vertical clearance on all
other highways should be at least 14 ft (4.25 m) over
the entire roadway width to which an allowance
should be added for resurfacing.
2.2.1.4 Sidewalks—Sidewalks, when used on bridges,
should be as wide as required by the controlling and
concerned public agencies, and preferably should be 5 ft
wide (1.5 m) but not less than 4 ft (1.25 m).
2.2.1.5 Curbs—There are two general classes of curbs.
These are “parapet” (nonmountable) and “vehicular mount-
able” curbs. Both may be designed with a gutter to form a
combination curb and gutter section. The minimum width of
curbs should be 9 in. (225 mm). Parapet curbs are relatively
high and steep faced. They should be designed to prevent the
vehicle from leaving the roadway. Their height varies, but it
343R-14 MANUAL OF CONCRETE PRACTICE
Fig. 2.2.1.5—Parapet curb and railing section.
should be at least 2 ft-3 in. (700 mm). When used with a
combination of curb and handrail, the height of the curb may
be reduced. Fig. 2.2.1.5 shows a parapet curb and railing
section which has demonstrated superior safety aspects, and
is presently used by state highway offices. Mountable curbs,
normally lower than 6 in. (150 mm), should not be used on
bridges except in special circumstances when they are used
in combination with sidewalks or median strips. The railing
and curb requirements, and the respective design loads, are
indicated in AASHTO HB-12. Curbs and sidewalks may
have vertical slits or other provisions for discontinuity, to
prevent them from participating in deck bending moments,
to reduce cracking of these elements.
2.2.1.6 Medians—On major highways the opposing
traffic flows should be separated by median strips. Wherever
possible, the lanes carrying opposing flows should be sepa-
rated completely into two distinct structures. However,
where width limitations force the utilization of traffic sepa-
rators (less than 4 ft wide) the following median sections
should be used:
a. Parapet sections 12 to 27 in. (300 to 700 mm) in height,
either integral or with a rail section, are recommended
in “Location, Section, and Maintenance of Highway
Traffic Barriers.”2-2 The bridge and approach parapets
should have the same section.
b. Low rolled curb sections or double curb units with
some form of paved surface in between are recom-
mended for low-speed roads in “Handbook of Highway
Safety Design and Operating Practices.”2-3
2.2.1.7 Railing—Railing should be provided at the edge
of the deck for the protection of traffic or pedestrians, or
both. Where pedestrian walkways are provided adjacent to
roadways, a traffic railing may be provided between the two,
with a pedestrian railing outside. Alternatively, a combina-
tion traffic-pedestrian railing may be used at the outside of
the pedestrian walkway. Railings may be made of concrete,
metal, timber or a combination of these materials. The
service loads for the design of traffic and pedestrian railings
are specified in AASHTO HB-12.
While the primary purpose of traffic railing is to contain
the average vehicle using the structure, consideration should
also be given to protection of the occupants of a vehicle in
collision with the railing, to protection of other vehicles near
the collision, to vehicles or pedestrians on roadways being
overcrossed, and to appearance and freedom of view from
passing vehicles. Traffic railings should be designed to
provide a smooth, continuous face of rail. Structural conti-
nuity in the rail members (including anchorage of ends) is
essential. The height of traffic railing should be no less than
2 ft-3 in. (700 mm) from the top of the roadway, or curb, to
the top of the upper rail members. Careful attention should
be given to the treatment of railing at the bridge ends.
Exposed rail ends and sharp changes in the geometry of the
railing should be avoided. The approach end of all guardrail
installations should be given special consideration to mini-
mize the hazard to the motorist. One method is to taper the
guardrail end off vertically away from the roadway so that
the end is buried as recommended in “Handbook of Highway
Safety Design and Operating Practices.”2-3
Railing components should be proportioned commensu-
rate with the type and volume of anticipated pedestrian
traffic, taking account of appearance, safety, and freedom of
view from passing vehicles. The minimum design for pedes-
trian railing should be simultaneous loads of 50 lb/ft (730
N/m) acting horizontally and vertically on each longitudinal
member. Posts should be designed for a horizontal load of 50
lb (225 N) times the distance between posts, acting at the
center of gravity of the upper rail.
The minimum height of pedestrian railing should be 3 ft-6
in. (1.1 m), measured from the top of the walkway to the top
of the upper rail member. Railings for walkways that are also
used as bicycle paths should have a height of 4 ft-6 in. (1.4 m).
2.2.1.8 Superelevation—Superelevation of the surface
of a bridge on a horizontal curve should be provided in
accordance with the applicable standard for the highway.
The superelevation should preferably not exceed 6 percent,
and never exceed 8 percent.
2.2.1.9 Surfacing—The road surface should be
constructed following recommendations in ACI 345.
2.2.1.10 Expansion joints—To provide for expansion
and contraction, joints should be provided at the expansion
ends of spans and at other points where they may be desir-
able. In humid climates and areas where freezing occurs,
joints should be sealed to prevent erosion and filling with
debris, or else open joints should be properly designed for
the disposal of water.
A State-of-the-Art Report on Joint Sealants is given in
ACI 504R.
2.2.2 Railway bridges
2.2.2.1 Railway classification—Rail lines are classified
by their purpose and function. Each type has its own require-
ments for design, construction, and maintenance.
2.2.2.2 Width—The width of the bridge should be based
on the clearance requirements of AREA Manual for Railway
Engineering, Chapter 28, Part 1, or to the standards of the
railway having jurisdiction.
2.2.2.3 Clearances—Minimum clearances should be in
accordance with the requirements of the railway having
 BRIDGE ANALYSIS AND DESIGN 343R-15

jurisdiction. Minimum clearances established by AREA are

indicated in Fig. 2.2.2.3.
2.2.2.4 Deck and waterproofing—All concrete decks
supporting a ballasted roadbed should be adequately drained
and waterproofed. The waterproofing should be in accor-
dance with the provisions outlined in AREA Manual of
Railway Engineering, Chapter 29.
2.2.2.5 Expansion joints—To provide for expansion and
contraction movement, deck expansion joints should be
provided at all expansion ends of spans and at other points
where they may be necessary. Apron plates, when used,
should be designed to span the joint and to prevent the accu-
mulation of debris on the bridge seats. When a waterproof
membrane is used, the detail should preclude the penetration
of water onto the expansion joint and bridge seat.
2.2.3 Aircraft runway bridges—The runway width,
length, clearances, and other requirements should conform
to the provisions of the Federal Aviation Agency or other air
service agency having jurisdiction.
2.2.4 Transit bridges—A transit bridge or guideway
differs from a conventional highway bridge in that it both
supports and guides an independent transit vehicle. Special
considerations are required in the design and construction to
attain the desired level of ride comfort. A State-of-the-Art
report for concrete guideways is given in ACI 358R.
2.2.5 Spans and profile
2.2.5.1 General—In addition to providing the proper
surface to carry the proposed traffic on the bridge, the struc-
ture should provide proper clearance for the facility being
crossed.
2.2.5.2 Stream and flood plain crossings—The bridge
should be long enough to provide the required waterway
opening below the high water elevation of the design flood.
The opening provided should be an effective opening, i.e., be
measured at a right angle to the stream centerline, furnish
adequate net opening, have adequate upstream and down-
stream transitional cleanouts, be vertically positioned be-
tween the stable flowline elevation and the correct frequency
highwater elevation, and be positioned horizontally to most
efficiently pass the design volume of water at the design flood
stage. Provision should be made for foreseeable natural
changes in channel location and, if necessary, channel realign-
ment should be made a part of the bridge construction project.
The bridge waterway opening and the roadway profile
together determine the adequacy of the system to pass
floods. The roadway and stream alignments determine the
effectiveness of the provided opening and they influence the
need for spur dikes, erosion protection, structure skew, and
pier locations. Detailed guidelines are given in AASHTO
HDG-7. Good practice usually dictates that the lowest eleva-
tion of the superstructure should not be lower than the all-
time record high water in the vicinity of the crossing and that Fig. 2.2.2.3—Clearance diagram for railroads.
an appropriate clearance be provided above the design high-
water elevation. The amount of clearance depends on the
type of debris that should pass under the bridge during floods such as lowering the approach embankment to permit over-
and the type of bridge superstructure. When costs of meeting topping. If any part of the bridge superstructure is below the
this requirement are excessive, consideration should be all-time record high water, it should be designed for stream
given to other means of accommodating the unusual floods, flow pressures and anchored accordingly.
343R-16 MANUAL OF CONCRETE PRACTICE
Requirements for stream crossings should be obtained
from the governmental agency having responsibility for the
stream being crossed. For further recommendations, see
Hydraulic Design of Bridges with Risk Analysis.2-4
2.2.5.3 Navigable stream crossings—Vertical clearance
requirements over navigation channels vary from 15 to 220
ft (4.5 to 67 m) and are measured above an elevation deter-
mined by the U.S. Coast Guard, or in Canada by Transport
Canada. Horizontal clearances and the location of the
opening depend upon the alignment of the stream upstream
and downstream of the bridge. In some cases auxiliary navi-
gation channels are required.
2.2.5.4 Highway crossings (Fig. 2.2.5.4)—The pier
columns or walls for grade separation structures should
generally be located a minimum of 30 ft (9 m) from the edges
of the through traffic lanes. Where the practical limits of
structure costs, type of structure, volume and design speed of
through traffic, span arrangement, skew, and terrain make
the 30-ft (9-m) offset impractical, the pier or wall may be
placed closer than 30 ft (9 m) and protected by the use of
guard rail or other barrier devices. The guard rail should be
independently supported with the roadway face at least 2 ft
(600 mm) from the face of pier or abutment.
The face of the guard rail or other device should be at least
2 ft (600 mm) outside the normal shoulder line.
A vertical clearance of not less than 14 ft (4.25 m) should
be provided between curbs, or if curbs are not used, over the
entire width that is available for traffic. Curbs, if used,
should match those of the approach roadway section.
2.2.5.5 Railway crossings (Fig. 2.2.5.5)—In addition to
the requirements shown in Fig. 2.2.5.5, it is good practice to
allow at least 6 in. (150 mm) for future track raise. In many
instances the railroad requires additional horizontal and
vertical clearance for operation of off-track equipment. Piers
located closer than 25 ft (7.5 m) from the track should meet
the requirements of AREA Manual for Railway Engi-
neering, Chapter 8, Subsection 2.1.5, to be of heavy
construction or to be protected by a reinforced concrete crash
wall extending 6 ft (2 m) above top of rail. In certain
instances where piers are adjacent to main line tracks, indi-
vidual railways may have more stringent requirements.
2.3—Esthetic considerations
A bridge should be designed to harmonize with its natural
surroundings and neighboring structures. The attractiveness
of a bridge is generally achieved by its shape and by the
proper proportioning of the superstructure and piers in rela-
tion to the span of the bridge and its surroundings. Color and
texture may be added for emphasis. Consideration should be
given to the appearance of the bridge from the driver’s or
passenger's point of view, as well as someone viewing it
from off the structure. A bibliography of books and articles
on bridge esthetics has been published (Reference 2-5) and
is available from the Transportation Research Board.
2.4—Economic considerations
2.4.1 Criteria for least cost—Least-cost criteria require
consideration of all the factors contributing to the cost of the
project. These include length and width of superstructure, type
of superstructure including deck, railings, walks, medians;
type of substructure including cofferdams, sheeting, and
bracing, approach roadways including embankment, retaining
walls and slope protection. Other factors such as special treat-
ment for the road or stream being spanned, and pier protection,
can also influence the least cost.
Each type of superstructure being considered has an
optimum span range where its use is very competitive. It
may, however, be used in spans outside that range and still
meet the least-cost criteria, because of the compensating
costs of other factors. One of the compensating factors often
is the substructure because its contribution to the cost of the
project is inversely proportional to the span length, while the
superstructure cost increases with the span length. Wherever
possible, consideration should be given to comparing bridge
layouts having different span arrangements. Elimination of a
costly river pier can usually justify a longer span.
Although this report stresses the design of the superstruc-
ture, the substructure of any bridge is a major component of
its cost and for that reason offers an almost equally great
potential for cost saving.
In the final analysis, however, true economy is measured
by the minimum annual cost or minimum capitalized cost for
its service life. Cost data on maintenance, repair or rehabili-
tation, and estimate of useful life are less easy to obtain, but
no study of least cost can be complete without their consid-
eration. For cost of bridges within flood plains, see
Hydraulic Design of Bridges with Risk Analysis.2-4
2.4.2 Alternative designs—The general statement that a
competent engineer can establish the most economical struc-
ture by studies ignores factors which influence costs over
which the engineer has no control. The economics of any
given industry cannot be exactly forecast. The time of adver-
tising most structures is not established at the time of design.
The reasons for preparing alternative designs are:
a. Increase competition by permitting several industries to
participate.
b. Make provisions to take advantage of the variations in
the economy of the construction industry.
c. To provide a yardstick whereby the various industries
can measure the advantage and disadvantage of their
competitive position. This results in industry improving
their procedures to reduce costs and eventually gives
additional savings to the owner.
d. To eliminate the intangible arguments by various
segments of industry that their material would have
resulted in a more economical structure.
e. Most important reason for alternative designs is that the
owner saves in the cost of the structure.
For more detailed information consult Alternate Bridge
Designs.2-6
2.4.3 Value engineering—In addition to economic pres-
sures, sociological pressures have focused more attention on
the impact that a project has on both natural and cultural
environments. Consequently, the bridge engineer is faced
with the necessity of identifying a continually growing list of
design parameters, along with the accompanying possibility
 BRIDGE ANALYSIS AND DESIGN 343R-17

Fig. 2.2.5.4—Clearance diagrams for underpasses.

343R-18 MANUAL OF CONCRETE PRACTICE
Fig. 2.2.5.5—Clearances for railway crossings.
of tradeoffs in the process of planning and designing. Selec-
tion of not only a suitable type of substructure and superstruc-
ture, but a suitable location with the consideration of all these
factors, can be very complex. A complete and objective result
can be accomplished only if an organized approach is adopted.
Value Engineering is one such system that can help engineers
obtain an optimum value for a project.
Value Engineering is an organized way of defining a
problem and creatively solving it. The Value Engineering
Job Plan has five steps: 1) information phase, 2) analysis
phase, 3) speculative phase, 4) evaluation phase, and 5)
implementation phase. The Job Plan encourages engineers to
search systematically, analyze objectively, and solve
creatively. Details of the Value Engineering method can be
obtained from Guidelines for Value Engineering2-7 or Value
Engineering for Highways.2-8
Value Engineering can be used at various stages of a
project, but the earlier the process is initiated, the greater the
possible benefits. This is graphically illustrated by Fig. 2.4.3,
taken from the book Value Engineering in Construction.2-9
Value Engineering when used only in the Value Engineering
Change Proposal (VECP) produces limited benefits because
only the low bidder on the “owner's design” is permitted to
submit a Value Engineering alternate. Its use is discussed in
Section 4.12 of this report.
2.5—Bridge types
Bridges may be categorized by the relative location of the
main structural elements to the surface on which the users
 BRIDGE ANALYSIS AND DESIGN 343R-19

travel, by the continuity or noncontinuity of the main

elements and by the type of the main elements.
2.5.1 Deck, half-through, and through types (see Fig.
2.5.1)—To insure pedestrian safety, bridges designed with
sidewalks should preferably permit an unobstructed view.
This requirement is satisfied with deck bridges where the
load-carrying elements of the superstructure are located
entirely below the traveled surface.
In rare cases, clearances may justify a half-through or
through-type structure when the difference between the
bridge deck elevation and the required clearance elevation is
small. In through-type structures, the main load-carrying
elements of the superstructure project above the traveled
surface a sufficient distance such that bracing of the main
load- carrying element can extend across the bridge. Thus Fig. 2.4.3—Potential Value Engineering savings during civil
the bridge user passes “through” the superstructure. works project life cycle.
In half-through structures, the main carrying elements are
braced by members attached to and cantilevering from the
deck framing system. The top flanges of half-through girders
or top chords of half-through trusses are much less stable
than those in deck and through structures. In the deck struc-
ture in particular, the roadway slab serves very effectively to
increase the lateral rigidity of the bridge. The projecting
elements of the half-through or through structures are very Fig. 2.5.1—Cross sections of bridge types.
susceptible to damage from vehicles and adequate protection
should be provided. These types also do not permit ready
widening of the deck in the future.
2.5.2 Simple, cantilever, and continuous span types (see
Fig. 2.5.2)—Concrete bridges may consist of simple, canti-
lever, or continuous spans. Continuous structures with vari-
able moments of inertia for slabs, stringers, and girder
systems require the least material. However, in the shorter
span range the labor cost of constructing variable sections
often offsets the material savings.
In the past the use of cantilever arms and suspended spans
rather than continuous structures resulted from fear of the
effect of differential settlement of supports. It should be
recognized that these effects can now be readily considered
by proper use of analytical methods and current knowledge
in soil mechanics and foundation engineering. Because of
the difficult problems of detailing and constructing the
bearing which forms the hinge at the end of the cantilever, its
use should be limited to special situations. It is also inadvis-
able to use hinges in areas subject to seismic loadings.
In the longer span range of slab, stringer, and girder-type
bridges, the use of continuous structures with variable
moments of inertia is strongly recommended. It should also be
recognized that the designer may sometimes take advantage of
the economy of a combination of bridge types. Short approach
spans of slab or box beam construction may be combined with Fig. 2.5.2—Elevations of bridge types.
a single main arch span or with long box girder spans.
2.5.3 Slab, stringer, and girder types—These types may
be either square or skew in plan and simply supported, canti- longitudinal reinforcement may be prestressed or nonpre-
levered, or continuous over supports. stressed. Transverse prestressing can be used for transverse
2.5.3.1 Slab type (see Fig. 2.5.3.1)—Slab-type concrete reinforcement in both cast-in-place slabs and those
bridges consist of solid or voided slabs which span between composed of longitudinal segments.
abutments and intermediate piers. The use of slab bridges 2.5.3.2 Stringer type (see Fig. 2.5.3.2)—The main struc-
should be considered for spans up to 80 ft (25 m). The main tural elements of this type of concrete bridge consist of a
343R-20 MANUAL OF CONCRETE PRACTICE
Fig. 2.5.3.1—Cross sections of slab bridges.
Fig. 2.5.3.2—Cross sections of stringer bridges.
Fig. 2.5.3.3—Cross sections of girder bridges.
Fig. 2.5.4—Elevations of rigid frame bridges.
series of parallel beams or stringers which may or may not
be connected with diaphragms. The stringers support a rein-
forced concrete roadway slab which is generally constructed
to act as the top flange of the stringer. Use of stringer bridges
should be considered for spans ranging from about 20 to 120 ft
(6 to 36 m). Stringers generally are spaced from 6 to 9 ft (1.8
to 2.7 m) on centers. Main reinforcement or prestressing is
located in the stringers. When the stringers are prestressed,
the concrete roadway slab may be reinforced or prestressed
in two directions.
2.5.3.3 Girder type (see Fig. 2.5.3.3)—This type of
concrete bridge consists of either longitudinal girders
carrying cross beams that in turn carry the roadway slab, or
longitudinal box girders whose bottom slab functions as the
bottom flange of the girder. In both types the top slab serves
the dual function of being the flange of the girder and the
roadway slab of the bridge.
Solid web girders—This type of bridge differs from the
stringer type in that the reinforced concrete roadway slab
spans longitudinally and is supported on cross beams spaced
8 to 12 ft apart (2.4 to 3.6 m). In general only two girders are
used. It is a feasible solution where large overhangs of the
deck are desirable, and where the depth of construction is not
critical. Spans of up to 180 ft (55 m) have been built.
Box girders—These girders may consist of a single cell for a
two-lane roadway, multiple cells for multiple-lane roadways, or
single or multiple cells with cantilever arms on both sides to
provide the necessary roadway width, and to reduce the
substructure cost and minimize right of way requirements.
Because of their superior torsional rigidity, box girders are
especially recommended for use on curves or where
torsional shearing stresses may be developed. This type of
concrete bridge may have economical span lengths from less
than 100 to about 700 ft (30 to 210 m). In the longer span
range, variable depth, variable moment of inertia, cantilever
construction, continuity design, pretensioned precast
elements, precast post-tensioned components, post-
tensioned construction, three-way prestressing for the whole
bridge assembly, and introduction of continuity after erec-
tion should all be considered.
2.5.4 Rigid-frame type (see Fig. 2.5.4)—Rigid-frame
bridges may be hinged or fixed, single or multiple spans. The
main structural elements are generally slabs, but may be
beams. Depending upon the roadway planning, each span
may accommodate from one lane with shoulders to as many
as four lanes with shoulders. A two-span structure can
accommodate roadways in each direction with a narrow
median. A three-span structure can accommodate roadways
in each direction with a wide median. Overhangs may be
introduced to span over embankment slopes and to reduce
the moment at the knees of the first legs of a rigid-frame
system. The base of legs or columns may be either hinged or
fixed. In either case, the hinge action or fixity should be
constructed to fit the design conditions. Feasible span
lengths are similar to slab or stringer-type bridges.
The bridge may be square or skew to suit the geometric or
hydraulic requirements. Where possible, large skews should
be avoided by changing the alignment of the supported
roadway or the obstruction being crossed. If this cannot be
done, a detailed analysis should be used in design of the
structure. In restricted locations, barrel-type bridges may be
used to dispense with abutments. Both composite construc-
tion and two-way reinforcing or two-way prestressing may
be applied to the slab-rigid-frame assembly.
2.5.5 Arch type—Arches generate large horizontal thrusts
at their abutments, and are therefore ideally suited for cross-
ings of deep gorges or ravines whose rock walls provide a
relatively unyielding support. If founded on less suitable
material, the deformation of the foundation should be taken
into account in the analysis. Arches are also best suited for
structures that have a sizable dead load-to-live load ratio.
Arch construction may be made of cast-in-place or precast
segmental elements, as well as being conventionally formed.
2.5.5.1 Spandrel or barrel arches (see Fig. 2.5.5.1)—
These arches are often fixed at the springing lines on the
abutments. Two-hinged and three-hinged arches are seldom
used. More often, single spans are used. In this type of
bridge, the spandrels act as retaining walls for earth fill
which is placed on top of the arch to form the subbase for the
roadway. It has distinct advantages for short spans, low rises,
and heavy live loads. Spans up to 200 ft (60 m) have been
used. However, in spans over 100 ft (30 m), the dead load
due to the earth fill may become excessive.
2.5.5.2 Ribbed or open-spandrel arches (see Fig.
2.5.5.2)—This type of bridge is suitable for long spans. Both
 BRIDGE ANALYSIS AND DESIGN 343R-21

fixed and two-hinged arches are common. The three-hinged

version is feasible but is seldom used.
Ribbed arches are better adapted to multiple spans than
barrel arches. The roadway deck is supported on columns
and cross beams. It may be a deck structure or half-through
structure. In the latter case, hangers should be used to carry
the roadway in the central portion of the arch. In this type of
bridge, these are usually two separate arch ribs, but there
may be more or the rib may be a solid slab, in which case the
columns and cross beams may be replaced by walls. Spans
up to 1000 ft (300 m) have been built. Fig. 2.5.5.1—Two span spandrel or barrel (earth filled) arch
2.5.5.3 Tied arches (see Fig. 2.5.5.3)—This type of arch bridge.
should be used where the foundation material is considered
inadequate to resist the arch thrust. Tied arches are always of
the through or half-through type with hangers to carry floor
beams. Both single-span and multiple-span tied arch bridges
have been built with spans of 100 to 400 ft (30 to 120 m).
2.5.5.4 Long-span arches—Ribbed arches with precast
box segments, post-tensioned during assembly, are suitable
for long spans. In these bridges, post-tensioned diaphragms
have been used. Fig. 2.5.5.2—Single span open spandrel or ribbed arch
2.5.5.5 Splayed arches or space frame—To achieve bridge.
greater stability and structural stiffness, through-type arches
may be inclined to each other to form a space frame. This
results in a splayed arch system.
For shorter spans, vertical cable suspenders may be used.
In the longer span range, diagonal-grid cable suspenders are
recommended for attaining greater stiffness.
The diagonal-grid concept may be further extended to the
floor system by having a grid of beams running diagonally to
the center line of the roadway. Fig. 2.5.5.3—Tied arch bridge.
2.5.6 Truss types—Although concrete truss bridges of
triangular configuration have been built, their use is not
recommended. The detail of reinforcement at a joint where
many members meet is very difficult. Formwork, centering,
and placing of concrete in sloping members is expensive.
2.5.6.1 Vierendeel truss (see Fig. 2.5.6.1)—While not
used extensively, the Vierendeel truss offers some esthetic
qualities, has simpler details because of the limited number
of members at a joint, is easier to form and place, and can be
precast or cast in place. If necessary, it can be erected by
cantilever method without false work.
Previously, a deterrent to the use of the Vierendeel truss Fig. 2.5.6.1—Vierendeel truss bridge.
was the difficulty in analyzing this highly indeterminate
structure, but computer programs are now available to
rapidly do this analysis. compression member. Cable-stayed bridges act as continuous
The use of inclined chords, particularly in the end panel, is girders on flexible supports, and offer the advantage of high
recommended. This inclination greatly reduces the bending rigidity and aerodynamic stability with a low level of
stresses. secondary stresses. Spans of up to 1500 ft (460 m) are feasible.
Because of the great depth needed for efficient use, the The proper design of the pylon and configuration of the
Vierendeel truss bridge is generally either a through or half- stays can add greatly to the esthetic appearance.
through type. It is also best suited for simple spans. Spans up 2.5.8 Suspension types (see Fig. 2.5.8)—In suspension
to 500 ft (150 m) have been built. bridges, intermediate vertical support is furnished by hangers
2.5.7 Cable-stayed types (see Fig. 2.5.7)—The feasible from a pair of large cables. The deck does not participate
span of concrete bridges can be greatly extended by the use except to span between hangers and to resist horizontal loads.
of supplementary supports. In cable-stayed bridges the Because dead load is a very important factor, high-
concrete deck, including the roadway slab and girders, acts strength lightweight concrete should be considered. Spans of
as a part of the support system, functioning as a horizontal 1600 to 1800 ft (489 to 550 m) have been built.
343R-22 MANUAL OF CONCRETE PRACTICE
Fig. 2.5.7—Cable stayed bridge.
Fig. 2.5.8—Suspension bridge.
2.6—Construction and erection considerations
In the design of a bridge, construction and erection consid-
erations may be of paramount importance in the selection of
the type of bridge to be built. Also, the experience of the
available contractors, the ability of local material suppliers
to furnish the specified materials, the skilled labor required
for a particular structure type, and the capacity of equipment
necessary for erection should be considered. The most
economical bridge design is one in which the total cost of
materials, labor, equipment, and maintenance is minimized.
2.6.1 Cast-in-place and precast concrete—The decision
to use or not to use precast concrete could be influenced by
the availability of existing precast plants within transport
distance. Precast concrete may be competitive in areas
without existing precasting plants when a large number of
similar components are required.
In large projects, a precasting plant located at the site should
be considered to see if it would prove more economical.
In general, precast concrete members, because of better
control of casting and curing processes, and because of the
ease of inspection and rejection of an improperly fabricated
member, are a better, more durable product.
For grade separation structures, if traffic problems are not
a controlling factor, cast-in-place structures are generally
more economical when the height of falsework is less than
30 to 40 ft (9 to 12 m) high.
2.6.2 Reinforced, partially prestressed, and prestressed—
This report covers use of reinforced concrete, partially
prestressed concrete, and fully prestressed concrete. The
possible use of pretensioning or post-tensioning should be
considered during the planning stage of a bridge project. In
many cases the greatest economy can be realized by
allowing the Contractor the option of using pretensioned,
post-tensioned, or a combination of both. In these cases, the
specifications should require submittal by the Contractor of
proper design data.
2.6.3 Composite construction—Integration of the deck
slab with the supporting floor system is covered by this
document. Floor systems consisting of stringers, floor
beams, or combinations can be used. Modular precast
concrete planks (prestressed or regular reinforced) may be
used as the bottom form for the deck slab between stringers.
Properly designed, these planks can be made composite with
the cast-in-place deck slab and the deck slab composite with
the stringers. Consideration should be given in the design to
construction loads supported prior to the cast-in-place
concrete attaining its design strength. For short spans within
the capacity of available handling equipment, the entire deck
span may be precast in one piece and made composite with
the cast-in-place slab.
2.6.4 Post-tensioned segmental construction—It is normal
practice to build concrete bridges in segments such as
precast I-beams with composite slabs or precast voided slabs
or box beams that are attached together. In the post-
tensioned segmental type, the individual member, box
girder, I-beam, or arch is installed in several longitudinal
segments and then post-tensioned together to form one
member.
2.6.4.1 Box girders—In general, the longer spans,
because of the need for greater and variable depths, have
been cast-in-place, while the shorter spans lend themselves
to constant depth precast units. It is customary to erect these
bridges by the cantilever method, avoiding the use of false-
work, but some have been erected using a limited amount of
falsework and placing the bridge by “pushing” the
completed segments into place from one end.
2.6.4.2 I-beams—Due to shipping limitations, the length
of precast prestressed I-beam stringer bridges is less than
100 ft (30 m). By precasting the I-beam in two or more
pieces and post-tensioning the pieces after erection, the
feasible span can be greatly increased.
2.6.4.3 Arches—Arches of all types may be constructed
of cast-in-place or precast segments. This method of
construction is most adaptable to long spans and spans where
centering for formwork is difficult to install. After
constructing the arch ribs by the segmental method, the span-
drel columns or suspenders and the roadway deck may be
constructed in a more conventional manner.
2.7—Legal considerations
2.7.1 Permits over navigable waterways—Preliminary
plans of a proposed bridge crossing any navigable waterway
should be filed with the Commandant, U.S. Coast Guard,
addressed to the appropriate District Commander, and with
other appropriate Governmental authority. A written permit
with reference to horizontal and vertical clearances under the
spans, and to the location of all river piers, should be
obtained. Special permit drawings, 8 x 101/2 in. (203 x 267
mm) in size, showing the pertinent data must be prepared.
These requirements are given in the latest issue of U.S. Coast
Guard Bridge Permit Application Guide of the appropriate
district. Since the Coast Guard districts do not follow state
boundaries, the address of the Coast Guard District having
jurisdiction can be obtained by contacting Chief, Office of
Navigation, U.S. Coast Guard (G-NBR), Washington, D.C.
20593, Phone - 202/755-7620.
 BRIDGE ANALYSIS AND DESIGN
In Canada such permit requirements can be obtained from
Chief NWPA, Program Division, Transport Canada, Coast
Guard, Ottawa, Ontario, Canada K1A ONT.
2.7.2 Environmental laws and national policy
a. The National Environmental Policy Act (NEPA) of
1969 (Public Law 91-190) requires that “all agencies of
the Federal Government... include in every recommen-
dation or report on major Federal actions significantly
affecting the quality of the human environment, a
detailed statement by the responsible official to
outline:
‘The environmental impact of the proposed action;
any adverse environmental effects which cannot be
avoided should the proposal be implemented; alterna-
tives to the proposed action; the relationship between
local short-term uses of man's environment and the
maintenance and enhancement of long-term produc-
tivity; and any irreversible and irretrievable commit-
ments of resources which would be involved in the
proposed action should it be implemented.’”
b. Section 4(f) of the Department of Transportation Act
(Public Law 89-670) declares that special effort should
be made to preserve the natural beauty of the country-
side and public park and recreational lands, wildlife
and waterfowl refuges, and historic sites. The Secre-
tary of Transportation...“shall not approve any
program or project which requires the use of any
publicly owned land from a public park, etc. unless (1)
there is no feasible and prudent alternative to the use of
such land, and (2) such program includes all possible
planning to minimize harm to such park, recreational
area, wildlife and waterfowl refuge, or historic site
resulting from such use.”
c. Historic preservation—The National Historic Preser-
vation Act of 1966 and Executive Order 11593, Protec-
tion & Enhancement of Cultural Environment, require
that Federal, or federally assisted projects must take
into account the project's effect on any district, site,
building, structure, or object that is included in the
National Register of Historic Places and give the Advi-
sory Council on Historic Preservation an opportunity
to comment on the undertaking. Further, federal plans
and programs should contribute to the preservation and
enhancement of sites, structures, and objects of histor-
ical, architectural, or archeological significance.
d. Clean Air Act—The impact of a bridge project on air
quality must be assessed, and the project must be
consistent with the state (air quality) implementation
plan.
The Federal Highway Administration's policies and
procedures for considering air quality impacts on
highway projects are contained in FHWA Program
Manual, Vol. 7, Chapter 7, Section 9 (Reference 2-10).
This manual and any state or local standards may be
used as a guide in determining the type of bridge projects
for which air quality impacts are a reasonable concern.
e. The Noise Control Act of 1972—This act establishes a
national policy to promote an environment free from
343R-23
noise that jeopardizes health and welfare. For bridge
projects where highway noise is a concern, FHWA
Program Manual, Vol. 7, Chapter 7, Section 3 (Refer-
ence 2-10) and/or state or local standards may serve as
a guide in evaluating and mitigating noise impacts.
f. The Federal Water Pollution Control Act Amendments
of 1972.—Section 401 requires that applicants for a
federal permit provide a water quality certificate by the
appropriate state or interstate agency. If there is no
applicable effluent limitation and no standards, the
state water quality certifying agency shall so certify. If
the state or interstate agency fails or refuses to act on a
request for certification within a reasonable length of
time (normally deemed to be 3 months, but not to
exceed 1 year) after receipt of request, the certification
requirements shall be waived. No permit will be
granted until certification has been obtained or waived,
or if certification has been denied.
Section 404 assigns to the Corps of Engineers the
responsibility for issuing permits for the discharge of
dredged or fill material. However, the environmental
documentation for a bridge project must contain an
analysis of the impact of any fill associated with that
project.
g. Fish and Wildlife Coordination Act—Section 2
requires that, “whenever the water of any stream or
other body of water are proposed or are authorized to
be ... controlled or modified for any purpose whatever
... by any department or agency of the United States, or
by any public or private agency under Federal Permit
or license, such department or agency shall first consult
with the United States Fish and Wildlife Service,
Department of the Interior, and with the head of the
agency exercising administration over the wildlife re-
sources of the particular state where ... (the facility is to
be constructed...).” The environmental documentation
for the bridge project should include an analysis of
probable impacts on fish and wildlife resources and an
analysis of any mitigative measures considered, and
adopted or rejected.
h. The Endangered Species Act of 1973—This act gener-
ally provides a program for the conservation, protec-
tion, reclamation, and propagation of selected species
of native fish, wildlife, and plants that are threatened
with extinction. Section 7 of this act provides that
federal agencies shall take “such actions necessary to
insure that actions authorized, funded or carried out by
them do not jeopardize the continued existence of such
endangered species and threatened species or result in
the destruction or modification of habitat of such
species.” The list of endangered and threatened
species, published by the Fish and Wildlife Service in
the Federal Register, shall be consulted to determine if
any species listed or their critical habitats may be
affected by the proposed project. Section 7 of this act
establishes a consultation procedure to avoid and miti-
gate impacts on listed species and their habitats.
343R-24 MANUAL OF CONCRETE PRACTICE
i. Water Bank Act—Section 2 of this Act declares that ...
“It is in the public interest to preserve, restore and
improve the wetlands of the Nation.” Bridge projects
must be planned, constructed, and operated to assure
protection, preservation, and enhancement of the
nation's wetlands to the fullest extent practicable.
Efforts should be made to consider alignments that
would avoid or minimize impacts on wetlands, as well
as design changes and construction and operation
measures, to avoid or minimize impacts.
j. Wild and Scenic Rivers Act—Section 7 of this act
provides generally, that no license shall be issued for
any water resources project where such project would
have a direct and adverse effect on a river or the values
for which such river was designated by this act. A
bridge is considered to be included in the term “water
resources project,” and a permit is a license.
k. Prime and Unique Farmlands—Impacts of bridge
projects on prime and unique farmlands, as designated
by the State Soil Conservation service (U.S.D.A.), must
be evaluated. Efforts should be made to assure that such
farmlands are not irreversibly converted to other uses
unless other national interests override the importance
of preservation or otherwise outweigh the environ-
mental benefits derived from their protection. Analysis
of the impact of a bridge project on any such land
SHALL be included in all environmental documents.
l. Executive Order 11988, Floodplain Management—
This Order sets forth directives to “avoid, to the extent
possible, the long and short term impacts associated
with the occupancy and modification of floodplains
and to avoid direct or indirect support of floodplain
development wherever there is a practicable alterna-
tive.” An analysis of a bridge project's effect on
hydraulics should be included in the environmental
documentation.
m. Relocation assistance—The Uniform Relocation and
Assistance and Real Property Acquisition Policies
Act of 1970 applies to projects where federal funds
are involved. If any federal funds are involved in a
bridge project, the environmental documents shall
show that relocated persons should be provided
decent, safe, and sanitary housing; that such housing
be available within a reasonable period of time
before persons are displaced; that such housing is
within the financial means of those displaced; and
that it is reasonably convenient to public services
and centers of employment.
n. Executive Order 11990, Protection of Wetlands—
Department of Transportation Policy is to avoid new
construction in a wetland unless: (a) there is no practi-
cable alternate to the construction, and (b) the
proposed project includes all practicable measures to
minimize harm to wetlands which may result from
such construction.
Wetlands are defined as lands either permanently or
intermittently covered or saturated with water. This
includes, but is not limited to, swamps, marshes, bogs,
sloughs, estuarine area, and shallow lakes and ponds
with emergent vegetation. Areas covered with water
for such a short time that there is no effect on moist-
soil vegetation are not included in the definition, nor
are the permanent waters of streams, reservoirs, and
deep lakes. The wetland ecosystem includes those
areas which affect or are affected by the wetland area
itself; e.g., adjacent uplands or regions up and down
stream. An activity may affect the wetlands indirectly
by impacting regions up or downstream from the
wetland, or by disturbing the water table of the area in
which the wetland lies.
2.7.3 Plans, specifications, and contracts—These engi-
neering documents together should define the work
expressly, clearly, thoroughly, and without possibility of
ambiguous interpretation. The plans should show all dimen-
sions of the finished structure, in necessary and sufficient
details to permit realization of the full intent of the design
and to facilitate the preparation of an accurate estimate of the
quantities of materials and costs. The plans should also state
which specification (e.g., AASHTO M77) was followed, the
loading the bridge was designed to carry, any other special
loading, the design strengths of materials (concrete, steel,
bearings), the allowable and design footing pressures, the
design method used (load factor or working stress), and the
design flood. The construction specifications and contracts
should also define construction methods, procedures, and
tolerances to insure workmanship, quality control, and appli-
cation of unit costs when stipulated under the contract. The
Contractor's responsibilities should be clearly defined in
detail, with everything expressly stated.
2.7.4 Construction inspection—The responsibilities of
construction inspection for concrete bridges should always
be clearly identified. Preferably, the owner should engage
the designer of the bridge to inspect its construction, to
review the contractor's procedures and falsework plans,
which should be submitted prior to construction.
RECOMMENDED REFERENCES
The documents of the various standards-producing organi-
zations referred to in this report are listed here with their
serial designation, including year of adoption or revision.
The documents listed were the latest effort at the time this
report was written. Since some of these documents are
revised frequently, generally in minor detail only, the user of
this report should check directly with the sponsoring group
if it is desired to refer to the latest revision.
American Association of State Highway and Transportation
Officials
HB-12 Standard Specifications for Highway Bridges,
13th Edition, 1983
DS-2 Design Standards—Interstate System, 1967
DSOF-3 Geometric Design Standards for Highways
Other Than Freeways, 1969
GD-2 A Policy on Geometric Design of Rural High-
ways, 1965
 BRIDGE ANALYSIS AND DESIGN 343R-25

GU-2 A Policy on Urban Highways and Arterial CITED REFERENCES

Streets, 1973 2-1. “Drainage of Highway Pavements,” Hydraulic Engi-
HDG-7 Hydraulic Analysis for the Location and Design neering Circular, No. 12, Mar. 1969, U.S. DOT, FHWA.
of Bridges, 1982 2-2. “Location, Section and Maintenance of Highway Traffic
GTB Guide for Selecting, Locating and Designing Barriers,” NCHRP Ref. #118—1971; TRB Washington, D.C.
Traffic Barriers, 1977 2-3. “Handbook of Highway Safety Design and Operating
HM-14 Standard Specifications for Transportation Practices,” U.S. DOT, FHWA, 1973.
Materials and Methods of Sampling and Testing 2-4. “Hydraulic Design of Bridge with Risk Analysis,”
USDOT FHWA-T5-80-226, Mar. 1980.
American Concrete Institute 2-5. “Bridge Aesthetics,” a bibliography compiled by
Martin P. Burke, Jr., P.E., published by Transportation
345-82 Standard Practice for Concrete Highway Bridge
Research Board.
Deck Construction
2-6. FHWA Technical Advisory T 5140.12, Dec. 4,
358R-80 State-of-the-Art Report on Concrete Guideways
1979,“Alternate Bridge Designs.”
504R-77 Guide to Joint Sealants for Concrete Structures 2-7. “Guidelines for Value Engineering,” Task Force #19,
SP-24 Models for Concrete Structures AASHTO-AGC-ARTBA Joint Cooperative Committee.
2-8. “Value Engineering for Highways,” Federal Highway
American Railway Engineering Association Manual for Administration, U.S. Department of Transportation
Railway Engineering 2-9. “Value Engineering in Construction,” Department of
Chapter 8—Concrete Structures and Foundations the Army, Office of the Chief of Engineers, Sept. 1974.
Chapter 28—Clearances 2-10. “Federal-Aid Highway Program Manual,” Federal
Chapter 29—Waterproofing Highway Administration, U.S. Department of Transportation.
343R-26 MANUAL OF CONCRETE PRACTICE
CHAPTER 3—MATERIALS
3.1—Introduction
The ultimate realization and performance of concrete
bridges depend upon well conceived and executed designs,
skilled construction, and the use of reliable materials. Mate-
rials used in construction of concrete bridges are presented
and/or referenced in this chapter. Also, recommended speci-
fications for materials acceptance, sampling, and testing are
presented and/or referenced.
3.2—Materials
3.2.1 Sources—Materials should be supplied from sources
approved before shipment. The basis for approval should be
the ability to produce materials of the quality and in the
quantity required. These approved sources should be used as
long as the materials continue to meet the requirements of
the specifications. It is recommended that materials be in
compliance with the standard specifications listed in the
following sections.
The sources of the materials should be identified and
contracts for their supply executed well in advance of the
time when concreting is expected to begin. Sufficient lead
time should be provided for the evaluating, sampling, and
testing of all material if sources have not been previously
approved by an appropriate agency, such as a State Depart-
ment of Transportation.
3.2.2 Specifications and standard practices—Material
specifications and tests for highway bridges should be in
compliance with current AASHTO “Standard Specifications
for Highway Bridges;” for railway bridges in compliance with
current AREA “Manual for Railway Engineering,” Chapter 8.
In addition, standard practices should be in accordance with
ACI guidelines and ASTM material specifications.
3.2.3 Admixtures—Admixtures are materials used to
modify the properties of concrete for a particular application.
Generally, admixtures are employed to increase strength,
improve workability and durability, and increase or decrease
the time of setting. To insure the desired product, care should
exercised in selection, evaluation, and methods of addition.
In evaluation, consideration should be given to the experi-
ence records of specific admixtures with concrete materials
commonly used in the area. ACI 212.1R and 212.2R provide
excellent resource information.
Air-entraining admixtures are used in bridge concrete to
improve the durability of concrete in freeze-thaw cycling,
particularly in the presence of deicing chemicals containing
chlorides. Recommended air contents for various nominal
coarse aggregate sizes are given in ACI 345. Specifications
for air-entraining admixtures are given in ASTM C 260.
Water-reducing admixtures conforming to ASTM C 494,
Type A or water-reducing and retarding admixtures
conforming to ASTM C 494, Type D allow for a reduction
in water-cement ratio and/or setting time for a given consis-
tency of concrete. Concrete workability is maintained while
strength is increased and concrete permeability reduced.
Water-reducing admixtures in combination with air entrain-
ment will produce greater resistance to chlorides in a freeze-
thaw environment. Additional economy will result where
specifications permit reducing cement content.
High-range water-reducing admixtures, which reduce the
quantity of mixing water by 12 percent or greater, are growing
in use by manufacturers of precast concrete. Increased work-
ability of the concrete mixture and accelerated compressive
strength gain substantially reduce the time required to achieve
stripping strengths. As in the case of conventional water-
reducing admixtures, these admixtures can behave differently
with different cements and temperatures. Two classes of high-
range water-reducing admixtures are specified, denoting
normal setting and retarding admixtures conforming to ASTM
C 494, Types F and G, respectively.
Accelerating admixtures are used to increase high-early
strength and decrease the setting time. Accelerators may be
specified to facilitate early form removal or cold weather
concreting.
Calcium chloride has been the most widely used acceler-
ator since it is very effective and relatively economical.
However, the use of calcium chloride in concrete promotes
corrosion of metals in contact with it, due to the presence of
chloride ions. Calcium chloride is not permitted where
galvanized metal stay-in-place forms are used, or for use in
prestressed concrete, or where dissimilar metals are
embedded. It should not be used in any elements of concrete
bridges/structures that may be exposed to additional chlo-
rides. Calcium chloride should not be used as an admixture
with lime-based slag cements, high-alumina cements, or
with super-sulfated cements. Consideration should be given
to obtaining the desired results without using an accelerating
admixture, either by use of a water-reducer or high-early
strength cement. AREA specifications do not permit the
addition of calcium chloride. If used, calcium chloride
should conform to ASTM D 98.
Calcium nitrite, although still under experimental evalua-
tion by some states, can be added to reinforced concrete to
inhibit corrosion and increase strength. Primary application
has been with precast, prestressed concrete box and girder
sections for bridges. The admixture should comply with
ASTM C 494, Type C.
There are three classes of mineral admixture conforming
to ASTM C 618: raw or calcined natural pozzolan (Class N),
and two classes of fly ash (Class F and Class C). Class F has
pozzolanic properties; Class C has some cementitious prop-
erties in addition to pozzolanic properties.
Mineral admixtures conforming to ASTM C 618 are used
in concrete to reduce the heat of hydration in mass concrete,
to improve the resistance of concrete to actions such as those
caused by reactive aggregates, and to conserve cement by
replacing a portion of the required cement except when high-
early strength is required.
3.2.4 Aggregates—Aggregates for concrete consist of fine
and coarse particles conforming to ASTM C 33. Both
 BRIDGE ANALYSIS AND DESIGN
AASHTO and AREA specifications require additional stan-
dard test methods. As a general guide, the maximum size of
the aggregate should not be larger than one-fifth of the
narrower dimension between sides of forms, one-third of the
depth of slabs, or two-thirds (AREA specifies one-half) of
the minimum clear spacing between reinforcing bars.
The sizes of the fine and coarse aggregates should be
reasonably well graded from coarse to fine, and the
maximum size of the coarse aggregate desired for specific
structures should be specified. The sizes of coarse aggre-
gates should conform to ASTM D 448. AREA specifications
modify ASTM D 448 for the sizes of coarse aggregates.
Lightweight aggregates may be specified in concrete for
structural elements. A prime consideration for bridge decks
is a reduction of dead load by approximately 25 percent.
Availability of lightweight aggregates at an economical
price may be a concern in some areas. Abrasion and dura-
bility should be evaluated when lightweight aggregate is
being considered for use in bridge decks or other exposed
locations. Lightweight aggregates, if required or permitted
by special provisions, should conform to ASTM C 330.
AREA specifications modify ASTM C 330.
The basic physical and chemical characteristics of aggre-
gate cannot be altered by processing, although the quantities
of certain deleterious particles can be reduced. Preparation
and handling affect such important aggregate properties as
gradation, uniformity of moisture content, cleanliness, and in
the case of crushed aggregate, particle shape, thereby having
an important influence on concrete quality. Frozen aggre-
gates or aggregates containing frozen lumps should be
thawed before use. Aggregates should have a reasonably
uniform moisture content when delivered to the mixer. Infor-
mation covering selection and application of aggregates for
concrete may be obtained from the report of ACI 221.
3.2.5 Cement—The type of cement should always be
specified.
Cement should conform to one of the specifications for
cement listed below:
a. Portland cement types as designated in ASTM C 150
b. Three shrinkage-compensating portland cements given
in ASTM C 845 designated as Types K, M, and S.
Recommended application is given in ACI 223.
c. Portland blast-furnace slag cement or portland-
pozzolan cement as given in ASTM C 595 for struc-
tural concrete.
The amount of cement used in the concrete should corre-
spond to the specified job mixture.
3.2.6 Water—Water for washing aggregates and for use with
cement in mortar or concrete should be clean and free from
deleterious amounts of oil, salt, sugar, acid, alkali, organic
materials, or other substances interfering with the required
strength, density, impermeability, and durability of concrete
and steel, or otherwise harmful to the finished product.
Nonpotable water should be tested and should meet the
suggested limits of AASHTO T 26. Nonpotable water
should not be used in concrete unless the following are satis-
fied: selection of concrete proportions shall be based on
concrete mixes using water from the same source; mortar test
343R-27
cubes made with nonpotable mixing water shall have 7-day
and 28-day strengths equal to at least 90 percent of strengths
of similar specimens made with potable water; strength test
comparison shall be made on mortars, identical except for the
mixing water, prepared and tested in accordance with ASTM
C 109. Cement paste setting should conform to ASTM C 191,
and concrete setting to ASTM C 403.
3.2.7 Selection of concrete proportions—Recommenda-
tions for selecting proportions for concrete are given in ACI
211.1. Provision is given for selecting and adjusting propor-
tions for normal weight concrete by the estimated weight
and/or the absolute volume methods.
Recommendations for lightweight concrete are given in ACI
211.2. Provision is made for proportioning and adjusting struc-
tural grade concrete containing lightweight aggregate.
Concrete ingredients and proportions should be selected to
meet the minimum requirements stated in the specifications
and contract documents. Field experience or laboratory trial
mixes are the preferred methods for selecting concrete
mixture proportions.
ACI 318 limits chloride ion content for corrosion protec-
tion, depending on member type. An initial evaluation can be
obtained by testing individual concrete ingredients for total
chloride ion content. Additional information is given in ACI
201.2R and ACI 222R on the effects of chlorides on the
corrosion of reinforcing steel. When coated reinforcement
steel is used, the preceding guidelines may be more restric-
tive than necessary.
3.2.8 Curing materials—Freshly cast concrete should be
protected from premature drying and excessive heat or cold.
To insure continued hydration at an optimum rate, the
concrete should be kept saturated by wet, membrane, or
steam curing for a given length of time. Preservation of the
concrete moisture content may be accomplished by using
burlap cloth, enclosure steam curing, liquid membrane-
forming compounds, or waterproof sheet materials.
Burlap cloth should be made of jute or kenaf conforming
to AASHTO M182. The cloth should remain wet for the
entire specified curing time.
The liquid membrane-forming compounds should be suit-
able for spraying on horizontal and vertical surfaces and
should conform to ASTM C 309. The compounds covered
by this specification are available in different colors and are
suitable for use as curing media for fresh concrete. They also
provide additional curing of the concrete after removal of
forms or after initial moist curing. The application of Type 2,
white pigmented compound is generally preferred because it
reduces the temperature rise in concrete exposed to the sun.
However, at times its application may lead to an unsightly
appearance since the compound does not wear off readily or
evenly. Type 1-D with fugitive dye may be found more
desirable for structural barriers or substructure work.
Waterproof sheet material should conform to ASTM C 71.
This material is placed on the surfaces of concrete for
minimizing moisture loss during the curing period, and in
the case of the white reflective-type materials for reducing
temperature rise in concrete exposed to the sun. Early
343R-28 MANUAL OF CONCRETE PRACTICE
application of this material may produce a smooth, glossy
surface finish which is undesirable for a bridge deck.
The requirements of curing practice as prescribed by ACI
308 should be followed.
3.2.9 Joint materials—ACI 504R should be consulted for
information on joint materials and construction.
3.2.9.1 Water stops—Water stops are used to prevent the
infiltration of debris and water at construction, contraction,
fixed and expansion joints. These are of metal, rubber, or
plastics. Where the function of the joint is to provide move-
ment, the water stops should be designed to permit such
movement and be of a shape and thickness that will accom-
modate the force effects anticipated in the water stop.
Spliced, welded, or soldered water stops should form contin-
uous watertight joints.
Metal water stops may be made of copper or stainless steel
strips. Copper water stops or flashings should conform to
ASTM B 152 Copper No. 11000, electrolytic tough pitch type,
light cold-rolled, soft annealed. Stainless steel water stops
should conform to ASTM A 167, Type 316 L; No. 2 D Finish.
Rubber water stops may be molded or extruded. Their
cross section should be uniform, free from porosity or other
defects. The material for water stop may be compounded
from natural rubber, synthetic rubber, or a blend of the two,
together with other compatible materials which will produce
a finished water stop conforming to the test requirements of
ASTM D 412, D 572, D 746, D 747, D 792, and D 2240.
Plastic water stops should be fabricated by an extrusion
process with a uniform cross section that is free from
porosity or other defects. The material used for the water
stop should be a homogeneous, elastomeric, plastic
compound of basic polyvinyl chloride and other material
which, after fabrication, should conform to the same ASTM
test requirements as rubber water stops.
3.2.9.2 Joint fillers—Expansion joints may be filled
with bituminous, cork, sponge rubber, or other approved
expansion joint filler material. Preformed joints that are of
bituminous type and are to be resilient should conform to
ASTM D 1751. Preformed joints that are of bituminous type
and are less than 1/2 in. thick, such as those used in parapet
joints over piers of continuous structures, should conform to
ASTM D 994.
Preformed joints that are nonbituminous should conform
to ASTM D 1752, Type I, for sponge rubber material when
resiliency is required and Type II for cork material when
resiliency is not required.
Preformed joints that do not require the resiliency
provided by rubber, and which are used for exposed
concrete, may be made of polystyrene material conforming
to AASHTO M 230. When the joint is not exposed, such as
for a column hinged at a footing, the preformed material may
consist of bituminous material conforming to ASTM D 994
or to cork material conforming to ASTM D 1752, Type II.
3.2.9.3 Joint sealants—Preformed expansion joints may
be sealed with cold-application-type sealer, a hot-poured
elastic type sealer, or an elastomeric-type joint seal. The
cold-application joint sealer should conform to ASTM D
1850. The hot-poured elastic type should conform to ASTM
D 1190, and the elastomeric type should conform to
AASHTO M 220.
3.2.9.4 Mechanically locked sealants—Numerous
proprietary mechanically locked sealants in compression
and/or tension are available. Prior to specifying sealants, test
data and performance information should be obtained from
the manufacturer, or an agency where testing or performance
has been monitored and evaluated. Details for application
and size selection are given in ACI 504R.
Mechanical locking of compression seals between
armoring improves performance since direct compression
alone cannot be relied on to keep the seal in place. For
example, multi-unit modular joint systems, employing a
number of transverse extruded steel sections having “locked
in” elastomeric seals, are available. To insure joint sealing
integrity, it is important to provide sufficient anchorage
between the armor and concrete anchorage to withstand
traffic impact. Good field inspection is necessary to insure
proper installation.
Strip seals are widely used in concrete bridge decks for
thermal movements up to 4 in. (100 mm). The preformed elas-
tomeric seal element is mechanically locked between armored
interfaces of extruded steel sections. During structure move-
ments, a preformed central hinge enables the strip seal gland
profile to fold between the steel extrusions. When properly
sized and installed, watertightness is insured by wedge-action
of the elastomeric lugs within the steel extrusions.
Steel used in the extrusions should conform to ASTM A 242,
A 36, or A 588. The elastomer used in the seal or gland element
should conform to the requirements of ASTM D 2628.
3.2.9.5 Steel joints—Due to the availability, effective-
ness, and range of movement of mechanically locked glands
or seals, open steel joints are becoming less attractive. Posi-
tive protection against joint leakage is required to prevent
deterioration of bridge bearings and substructure units. Steel
joints can be made watertight by specifying a neoprene
trough. However, past experience indicates that regular
maintenance is required to keep the trough free of debris.
Steel joints are either a sliding plate or finger type. The
sliding is generally limited to thermal movements up to 4 in.
(100 mm) and the steel finger type up to 8 in. (200 mm). The
sliding plate and finger joints should be examined for
possible warpage before installation by laying the plates
together, loose, or on a flat surface. The steel should conform
to ASTM A 36 or A 588. Neoprene, if used, should conform
to ASTM D 412 and D 2240.
3.2.10 Bearings—The function of bridge bearings is to
transfer loads from the superstructure to the substructure.
Also, bearings should accommodate rotational movements
of the superstructure elements. Bearing guidelines are in the
process of being written by ACI Committee 554.
Expansion bearings should permit both longitudinal and
rotational movements while transferring lateral loads such as
wind. The coefficient of friction on mating surfaces should
be as low as practical; high frictional forces may increase the
cost of substructure units.
Fixed bearings should allow rotational movement while
transferring both lateral and longitudinal loads to the
 BRIDGE ANALYSIS AND DESIGN
substructure units. Fixed bearings tie the superstructure to
the substructure, thus preventing the bridge from potential
down-grade translation.
Several types of bearings are available; some of the more
common types are as follows:
3.2.10.1 Elastomeric bearings—Elastomeric bearings
are either molded, single-unit laminated pads with integral
layers of nonelastic shims or nonlaminated, molded, or
extruded pads. The elastomer should be natural rubber or
neoprene of Grade 55 + 5, durometer hardness material con-
forming to the tests of ASTM D 2240, D 412, D 573, D 395,
D 1149, D 429, and D 746, Procedure B. Experience indi-
cates that steel laminates are superior to other nonelastic
laminates. Laminates should be rolled, mild steel sheets
conforming to ASTM A 570, A 36, or A 611, Grade D. All
components of laminated bearings should be molded together
in an integral unit and all laminated edges should be covered
by a minimum 1/8-in. (3-mm) thickness of elastomer.
3.2.10.2 PTFE slide bearings—Polytetrafluroethylene
(PTFE) slide bearings are self-lubricating and can be bonded
to a rigid back-up material capable of resisting horizontal
shear and bending stresses. Expansion bearings of PTFE are
not recommended without providing an elastomer or rocker
plate to accommodate rotation. Stainless steel or other
equally corrosive-resistant material should be used for a
smooth, low-friction mating surface. Stainless steel and
unfilled PTFE made with virgin TFE resin should conform
to ASTM A 240, Type 316 and D 1457, respectively.
3.2.10.3 Steel bearings—Steel bearings are either roller,
rocker, sliding, or large built-up rocker types. Steel bearings
are fabricated from materials conforming to ASTM A 36, A
572, or A 588. All structural steel bearing plates should be
flat-rolled with smooth surfaces free of warp, having edges
straight and vertical. If lubricated bronze plates are required,
they should conform to ASTM B 22, Alloy 911 or B 100,
Copper Alloy 510.
On painted structures, the upper 6 in. (150 mm) of anchor
bolts, nuts, and washers should have a protective coating.
If specified, galvanizing should conform to ASTM A 153
or B 633.
3.2.10.4 Pot bearings—Pot bearings consist of a circular
steel piston and cylinder which confine an elastomeric pad.
The elastomer is prevented from bulging by the pot and acts
similarly to a fluid under pressure. The pot bearing serves as
an economical alternate to built-up steel bearings for higher
load applications. These bearings allow rotational movement
either with or without lateral and longitudinal movement.
Polyether urethane elastomer of pure virgin material
conforming to ASTM D 2240, D 412, or D 395 is recom-
mended for the rotational element. Longitudinal movement
can be attained by incorporating a PTFE slide bearing. The
steel cylinder and piston should preferably be machined
from a single piece of steel conforming to ASTM A 588.
3.2.10.5 Shear inhibited disc bearings—Shear inhibited
disc bearings consist of a load-bearing and rotational disc of
polyether urethane enclosed between upper and lower steel
bearing plates equipped with an internal shear restriction pin.
Expansion is accommodated by using a recessed PTFE on
343R-29
the upper half of the top bearing plate. The PTFE surface
supports an upper steel plate having a stainless steel surface.
The upper steel plate is fitted with guide bars to restrict
lateral movement.
The rotational element should be molded from polyether
urethane conforming to ASTM D 412, D 395, and D 2240.
Steel and stainless steel should conform to ASTM A 36 and
ASTM A 167, Type 316, respectively, unless otherwise
specified in the contract documents. PTFE should be made
from pure virgin unfilled teflon resin conforming to ASTM
D 638 and D 792.
3.2.11 Metal reinforcement
3.2.11.1 Reinforcing bars—All reinforcing bars should
be deformed except plain bars may be used for spirals or for
dowels at expansion or contraction joints. Reinforcing bars
should be the grades required by the contract documents, and
should conform to one of the following specifications:
a. ASTM A 615
b. ASTM A 616, including Supplement SI
c. ASTM A 617
d. ASTM A 706
Billet-steel reinforcing bars conforming to ASTM A 615,
Grade 60 [minimum yield strength 60 ksi (414 MPa)], are the
most widely used type and grade. The current edition of
ASTM A 615 covers bar sizes #3 through #11, #14, and #18.
ASTM A 615M covers metric bar sizes #10 through #55.
When important or extensive welding is required, or when
more bendability and controlled ductility are required as a
seismic-resistant design, use of low-alloy reinforcing bars
conforming to ASTM A 706 should be considered. Before
specifying A 706 reinforcing bars, however, local avail-
ability should be investigated.
3.2.11.2 Coated reinforcing bars—When coated rein-
forcing bars are required as a corrosion-protection system,
the engineer should specify whether the bars are to be zinc-
coated (galvanized) or epoxy-coated. The reinforcing bars to
be coated should conform to the specifications listed in
Section 3.2.11.1.
a. Zinc-coated (galvanized) reinforcing bars should
conform to ASTM A 767. Supplementary require-
ment S1 should apply when fabrication of the galva-
nization includes cutting. Supplementary requirement
S2 should apply when fabrication after galvanization
includes bending.
b. Epoxy-coated reinforcing bars should conform to
ASTM A 775.
c. Repair of damaged zinc coating, when required,
should use a zinc-rich formulation conforming to
ASTM A 767. Repair should be done in accordance
with the material manufacturer's recommendations.
d. Repair of damaged epoxy-coating, when required,
should use a patching material conforming to ASTM
A 775. Repair should be done in accordance with the
material manufacturer’s recommendations.
For zinc-coated reinforcing bars (galvanized) in accor-
dance with ASTM A 767, the engineer should specify class
of coating, whether galvanization is to be performed before
or after fabrication, and if supplementary requirement S3
343R-30 MANUAL OF CONCRETE PRACTICE
applies. If the bars are to be galvanized after fabrication, the
engineer should indicate which bars require special finished
bend diameters, such as the smaller bar sizes used for stirrups
and ties. All other reinforcement and embedded steel items
in contact with or in close proximity to galvanized rein-
forcing bars should be galvanized to prevent a possible reac-
tion of dissimilar metals.
On projects using both coated and uncoated bars, the engineer
should clearly indicate on the plans which bars are to be coated.
3.2.11.3 Bar mats—Bar mat reinforcement consists of
two layers of deformed reinforcing bars assembled at right
angles to each other, and clipped or welded at the bar inter-
sections to form a grid. The reinforcing bars used in mats
should conform to the specifications listed in Section
3.2.11.1. Whether clipped or welded mats are required
depends on the size of the mat and the rigidity required for
preserving the shape of the mat during handling. Clipping
the bars should be adopted whenever possible, as welding
decreases the fatigue strength of the bar steel because of the
stress concentration effect of the weld. Bar mats should
conform to ASTM A 184.
Clipped bar mats may be fabricated from zinc-coated (galva-
nized) reinforcing bars. Metal clips should be zinc-coated
(galvanized). Nonmetallic clips may be used. Coating damage
at the clipped intersections should be repaired in accordance
with recommendations given in Section 3.2.11.2(c).
Clipped bar mats may be fabricated from epoxy-coated
reinforcing bars. Metal clips should be epoxy-coated.
Nonmetallic clips may be used. Coating damage at the
clipped intersections should be repaired in accordance with
the recommendations given in Section 3.2.11.2(d).
3.2.11.4 Wire—Wire should be plain or deformed wire
as indicated in the contract documents. Spirals may be plain
wire. For wire with a specified yield strength fy exceeding
60,000 psi (414 MPa), fy should be the stress corresponding
to a strain of 0.35 percent.
Plain wire should conform to ASTM A 82.
Deformed wire should conform to ASTM A 496, size D4
and larger.
3.2.11.5 Welded wire fabric—Welded wire fabric is
composed of cold-drawn steel wires, which may or may not
be galvanized, and fabricated into sheets by the process of
electrical-resistance welding. The finished material consists
essentially of a series of longitudinal and transverse wires
arranged substantially at right angles to each other, and
welded together at all points of intersection. The use of this
material for concrete reinforcement should preferably be at a
location where the application of the loading is static or
nearly so. If the application of the loading is highly cyclical,
the fatigue strength of the wire fabric should be considered
when determining the sizes of the wires. For fatigue consid-
erations reference should be made to ACI 215R. The steel
wires for the fabric may be either plain or deformed. The engi-
neer should specify the size and wire spacing, and whether the
fabric is to be plain or deformed. For wire with a specified
yield strength fy exceeding 60,000 psi (414 MPa), fy should be
the stress corresponding to a strain of 0.35 percent.
Welded plain wire fabric should conform to ASTM A 185,
and welded intersections should not be spaced farther apart
than 12 in. (300 mm) in the direction of the primary flexural
reinforcement.
Welded deformed wire fabric should conform to ASTM A
497, and wire should not be smaller than size D4, and welding
intersections should not be spaced farther apart than 16 in. (400
mm) in the direction of the primary flexural reinforcement.
Welded wire fabric made from a combination of either
deformed or plain wires should conform to ASTM A 497 with
the same exceptions previously listed for deformed fabric.
3.2.11.6 Prestressing tendons—Strands, wire, or high-
strength bars are used for tendons in prestressed concrete.
a. Strands. Two grades of seven-wire, uncoated, stress-
relieved steel strand are generally used in prestressed
concrete construction. These are Grades 250 and 270,
which have minimum ultimate strengths of 250,000 and
270,000 psi (172 and 186 MPa), respectively. The
strands should conform to ASTM A 416. Supplement 1
of ASTM A 416 describes low-relaxation strand and
relaxation testing. Recommended test procedures for
stress relaxation are given in ASTM E 328.
b. Wire. Two types of uncoated stress-relieved round
high-carbon steel wires commonly used in prestressed
linear concrete construction are: Type BA wire for
applications in which cold-end deformation is used for
anchoring purposes (Button Anchorage) and Type WA
wire for applications in which the ends are anchored by
wedges and no cold-end deformation of the wire is
involved (Wedge Anchorage).
Post-Tensioning systems commonly used are
described in Post-Tensioning Manual by the Post-
Tensioning Institute. Types WA and BA wire should
conform to ASTM A 421. Supplement 1 of ASTM A
421 describes low-relaxation strand and relaxation
testing. Recommended test procedures for stress relax-
ation are given in ASTM E 328.
c. Uncoated high-strength bars for prestressed concrete
construction should conform to ASTM A 722. The
specification covers two types of bars: Type I (plain)
and Type II (deformed).
3.2.11.7 Structural steel, steel pipe, or tubing—Structural
steel used with reinforcing bars in composite columns should
conform to ASTM A 36, A 242, A 441, A 572, or A 588.
Steel pipe or tubing for composite columns should
conform to ASTM A 53 (Grade B), A 500, or A 501.
3.2.12 Accessories
3.2.12.1 Bar supports—All reinforcement should be
supported and fastened together to prevent displacement
before and during casting of concrete. Bar supports may
consist of concrete, metal, plastic, or other acceptable mate-
rials. Details and recommended practices for bar supports
are given in ACI 315.
Standardized, factory-made steel wire bar supports are the
most widely used type. Where the concrete surface will be
exposed to the weather in the finished structure, the portions
of bar supports near the surface should be noncorrosive or
protected against corrosion.
 BRIDGE ANALYSIS AND DESIGN
Bar supports for supporting zinc-coated (galvanized) or
epoxy-coated reinforcing bars should be in accordance with the
recommendations given in Section 13.11 of this document.
3.2.12.2 Side form spacers—Side form spacers, if
needed, are placed against vertical forms to maintain
prescribed clear concrete cover and position of the vertical
reinforcing bars. The need for side form spacers is deter-
mined by the proportions of the form, the arrangement and
placing of the reinforcing bars, the form material and forming
systems used, and the exposure of the surface to weather
and/or deleterious materials. In situations where spacers are
needed, various devices can be used such as double-headed
nails, form ties, slab or beam bolsters (wire bar supports),
precast concrete blocks, proprietary all-plastic shapes, etc.
The engineer should specify the requirements for side
form spacers including material, type, spacing, and location
where required.
Section 1.9 of AREA Manual, Chapter 8, specifies that at
all vertical formed surfaces that will be exposed to the
weather in the finished structure, side form spacers spaced
no further than 4 ft on center shall be provided. Spacers and
all other accessories within 1/2 in. of the concrete surface
shall be noncorrosive or protected against corrosion.
3.2.12.3 Tie wire—The tie wire used to fasten rein-
forcing bars should be black annealed steel wire, 16-gage
(1.6- mm diameter) or heavier. Tie wire for fastening zinc-
coated (galvanized) or epoxy-coated reinforcing bars should
be in accordance with the recommendations given in Section
13.11 of this document.
3.2.12.4 Bar splicing material—When required or
permitted, welded splices or mechanical connections may be
used to splice reinforcing bars.
All welding of reinforcing bars should conform to “Struc-
tural Welding Code—Reinforcing Steel” AWS D1.4. Rein-
forcing bars to be welded should be indicated on the
drawings and the welding procedure to be used should be
specified. Except for ASTM A 706 bars, the engineer should
specify if any more stringent requirements for chemical
composition of reinforcing bars than those contained in the
referenced ASTM specifications are desired, i.e., the chem-
ical composition necessary to conform to the welding proce-
dures specified in AWS D1.4.
Proprietary splice devices are available for making
mechanical connections. Performance information and test
data should be secured from manufacturers. Descriptions of
the physical features and installation procedures for selected
splice devices are given in ACI 439.3R.
3.2.12.5 Tensioning tendon components—Anchorages,
couplers, and splices for post-tensioned reinforcement
should develop the required nominal strength of the tendons,
without exceeding the anticipated set. Anchorages for
bonded tendons should develop at least 90 percent of the
specified ultimate strength of the prestressing steel when
tested in an unbonded condition, without exceeding the
anticipated set. Couplers and splices should be placed in areas
approved by the engineer and enclosed in housings long enough
to permit the necessary movements. They should not be used at
points of sharp curvature and should be staggered.
343R-31
Performance specifications for single unbonded strand
post-tensioning tendons are provided in the Post-Tensioning
Manual of the Post-Tensioning Institute. Unbonded tendon
anchorages should be subjected to the following additional
requirements as recommended in ACI 423.3R.
a. Static tests—When an assembly consisting of the
tendon and fittings is statically loaded, it should meet
the requirements set forth in the ASTM tendon material
specifications for yield strength, ultimate strength, and
minimum elongation. If minimum elongation at
rupture is not stated by ASTM specifications, the elon-
gation of the assembly should not be less than 3 percent
measured on not less than a 10-ft (3-m) gage length.
b. Cyclic tests—The test assembly should withstand,
without failure, 500,000 complete cycles ranging between
60 and 66 percent of the specified ultimate strength.
c. When used in structures subjected to earthquake load-
ings, the test assembly should withstand, without
failure, a minimum of 50 complete cycles of loading
corresponding to the following percentages of the mini-
mums specified ultimate strength in ksi (MPa)
60 + 2000/ln + 100
(60 + 610/ln + 30.5)
where ln is the length of the tendon in feet (meters).
d) Kinking and possible notching effects by the
anchorage should be avoided, particularly where
tendon assemblies will be subjected to repetitive or
seismic loadings.
3.2.13 Appurtenances
3.2.13.1 General—Materials in contact with and
partially embedded in concrete should be so constituted as to
not be injurious to the concrete.
3.2.13.2 Forms—Metal forms that are to remain in place
should be zinc-coated galvanized, both for appearance and
durability. Material for metal forms should conform to
ASTM A 446 Coating Designation G165. Any exposed form
metal where the zinc coating has been damaged should be
thoroughly cleaned and wire brushed, then painted with two
coats of zinc oxide-zinc dust primer, Federal Specification
TT-P-641d, Type II, no color added.
Details of formwork, design criteria and descriptions of
common types are shown in ACI SP-4.
3.2.13.3 Form coatings—Oil or other types of coating
used on forms to prevent sticking of the concrete should not
cause softening or permanent staining of the concrete
surface, nor should it interfere with any curing process which
might be used after form removal. Surfaces of forms made
with lumber containing excessive tannin or other organic
substance sufficient to cause softening of the concrete
surface should be treated with whitewash or limewater prior
to applying the form oil coating. Shellac, lacquers, and
compounded petroleum oils are commercially available for
form coating. The manufacturer of the coating should certify
that the product will not be deleterious to concrete.
3.2.13.4 Galvanized materials—When steel bolts, nuts,
and washers for anchoring railing posts, luminaires, pedes-
343R-32 MANUAL OF CONCRETE PRACTICE
trian chain link fences, ladders, stairways, joint dams, bear-
ings, as well as junction boxes, conduits and fittings, and
exposed steel inserts are galvanized, the protective zinc
covering should conform to ASTM A 153. When door
frames and doors provided for access to cells of box girders,
operator machinery, pump rooms, and floor drains fabricated
of structural steel are galvanized, the protective zinc
covering should conform to ASTM A 123.
3.2.13.5 Cast iron and stainless steel—The material
for cast iron hardware such as scuppers and drains,
embedded in concrete, should preferably conform to
ASTM A 48. When steel fasteners are embedded in
concrete and are subjected to alternate wetting and drying,
the material should preferably consist of stainless steel
conforming to ASTM A 276, Type 316.
3.2.14 Storage of materials
3.2.14.1 Cement—The handling and storage of cement
should be such as to prevent its deterioration or the intrusion
of foreign matter and moisture. The recommendations given
in ACI 304 should be followed.
3.2.14.2 Aggregates—Aggregates should be handled
and stored in such a manner as to minimize segregation,
degradation, contamination, or mixing of different kinds and
sizes. When specified, the coarse aggregate should be sepa-
rated into two or more sizes to secure greater uniformity of
the concrete mixture. Different sizes of aggregate should be
stored in separate stock piles sufficiently removed from each
other to prevent the material at the edges of the piles from
becoming intermixed.
3.2.14.3 Metal reinforcement—Reinforcement should
be stored above the surface of the ground and should be
protected from mechanical injury and surface deterioration
caused by exposure to conditions producing rust.
Prestressing tendons should be protected at all times
against physical damage and corrosion, from manufacture to
either grouting or encasing in concrete. Prestressing tendons
should be packaged in containers or shipping forms for the
protection of the steel against physical damage and corrosion
during shipping and storage. A corrosion inhibitor which
prevents rust or other results of corrosion should be placed in
the package or form, or the tendons should be precoated with
water-soluble oil. The corrosion inhibitor should have no
deleterious effect on the steel or concrete or bond strength of
steel to concrete. Care should be taken in the storage of
prestressing tendons to prevent galvanic action.
3.3—Properties
Concrete is not a homogeneous material. Its properties
cannot be predicted with accuracy. They vary not only with
the ingredients that make up concrete, but with the method
of mixing, placing, and even the loading. The design engi-
neer has only partial control of these variables, so it is neces-
sary to understand them, the probable range of their
variation, and their effect on the bridge structure. Only with
a thorough understanding can the engineer design and detail
a structure that functions properly.
3.3.1 Compressive strength—The design engineer speci-
fies the strength of concrete needed to insure the structural
adequacy. In the past it was always considered that the
stronger the concrete the better. Today's designer should
realize that some of the other properties that are directly
related to the strength of the concrete may not make it desir-
able to allow concrete significantly stronger than specified.
3.3.2 Tensile strength—In determining the strength of a
structure, the tensile strength of concrete is not directly
considered. However, it can be a significant factor in deflec-
tion computations.
3.3.3 Modulus of elasticity and Poisson’s ratio—The
modulus of elasticity for most bridge structures, using normal
weight concrete [(wc between 90 and 150 pcf (1400 and
2400 kg/m3)], can be assumed to be that given in ACI 318-83
Ec = wc1.5 33 fc′ psi (Ec = wc1.5 0.043 fc′ MPa)
However, in ACI 363R-84, attention is called to the signif-
icant difference between actual tests made on many high-
strength concretes and the ACI 318 formula.
If the structure being designed is of a significant size or if
deflections are of significant importance, the designer should
consider requiring tests to be made of the actual design
mixture that would be used in the structure. Using the results
of these tests in predicting deflections will be cost-effective,
especially considering the time and effort necessary to
correct the effects of improper camber and incorrectly
predicted deflections.
Poisson’s ratio may be assumed to be 0.2.
3.3.4 Creep—The inelastic strain of concrete under
compressive stresses is generally a beneficial property. In
this manner an indeterminate structure is able to adjust the
dead load stresses to more efficiently utilize the inherent
strength of the structure. However, when deflections due to
creep become significant, it should be considered in the
design. Creep decreases as the concrete ages, so one way to
partially control creep is to prevent early loading of the struc-
ture or member. Creep also decreases with an increase in
compressive reinforcement.
Creep can be as significant a property as the modulus of
elasticity and should always be considered by the designer.
Reference information for creep and shrinkage data is given
in ACI 209R, Chapter 5.
3.3.5 Shrinkage—During the drying process concrete
shrinks. The earlier the age that the concrete is dried, the
greater the shrinkage. Since concrete is rarely exposed
uniformly, the shrinkage of concrete is not uniform and
internal shrinkage stresses are induced. Also, concrete
sections are not restrained equally. This is particularly true at
construction joints where fresh concrete is restrained by the
previously cast concrete. The location of construction joints,
the rate of concrete placement, and the concrete placing
sequence can have a significant influence on the size and
number of shrinkage-related cracks. The designer should be
aware of the effects of shrinkage and creep and detail the
structure to minimize the adverse effects.
The shrinkage coefficient for normal weight concrete
should be established considering the local climatological
conditions and the construction procedures. However, it
 BRIDGE ANALYSIS AND DESIGN 343R-33

should not be taken as less than an equivalent strain of comparing the two specifications. In many cases, however,
0.0002. Shrinkage coefficients for lightweight concrete there are significant differences between them.
should be determined for the type of aggregate used.
3.3.6 Thermal coefficient—Concrete and steel have 3.4.1 ACI guidelines and standard practices
approximately the same thermal coefficient of expansion so 116R Cement and Concrete Terminology
that within the normal ranges the two materials act together. 201.2R Guide to Durable Concrete
If temperature changes were unrestrained and uniform, there 209R Predictions of Creep, Shrinkage and Tempera-
would be no need for concern. However, most concrete ture Effects in Concrete Structures
bridges have some degree of restraint, whether built-in or 211.1 Standard Practice for Selecting Proportions for
resulting from poor maintenance, deterioration of bearings Normal, Heavyweight, and Mass Concrete
and expansion details, or horizontal movements of supports. 212.1R Admixtures for Concrete
In addition, concrete is not a good thermal conductor. This 212.2R Guide for Use of Admixtures in Concrete
lack of conductivity results in significant temperature differ- 213R Guide for Structural Lightweight Aggregate
entials, particularly between the upper and lower surfaces of Concrete
the bridge. Thus stresses due to temperature effects are 215R Considerations for Design of Concrete Struc-
induced that can be of great significance. tures Subject to Fatigue Loading
Unlike creep and shrinkage, temperature effects are both 221R Guide for the Use of Normal Weight Aggre-
shortening and lengthening; but because temperature short- gates in Concrete
ening is additive to creep and shrinkage, the shortening 222R Corrosion of Metals in Concrete
effect is generally more critical. Unlike shrinkage and dead 223 Standard Practice for the Use of Shrinkage-
load stresses, the temperature effect is not a sustained load, Compensating Concrete
and thus cannot be alleviated by creep. A method of calcu- 304 Recommended Practice for Measuring,
lating temperature differentials is given in Section 5.4. Mixing, Transporting, and Placing Concrete
The thermal coefficient for normal weight concrete may 308 Standard Practice for Curing Concrete
be taken as 0.000006 F (0.000011 C). Thermal coefficients 315 Details and Detailing of Concrete Reinforcement
for lightweight concrete should be determined for the type of 318 Building Code Requirements for Structural
aggregate used. Concrete
3.3.7 State-of-the-art—For structures of greater than usual 345 Standard Practice for Concrete Highway
size, significance, or degree of complexity, the designer Bridge Deck Construction
should consult the latest publications of ACI Committees: 363R Report on High-Strength Concrete
209 Creep and Shrinkage, 363 High Strength Concrete, and 423.3R Recommendations for Concrete Members
213 Lightweight Aggregates and Concrete. Currently these Prestressed with Unbonded Tendons
include SP-9, SP-27, SP-29, SP-76, ACI 209R, ACI 213R, 439.3R Mechanical Connections of Reinforcing Bars
and ACI 363R. 504R Guide to Joint Sealants for Concrete Structures
3.3.8 Reinforcement properties—The modulus of elas- 554 Committee Correspondence on Bearing Systems
ticity of deformed steel reinforcing bars or welded wire SP-4 Formwork for Concrete
fabric may be taken as 29,000,000 psi (200,000 MPa). The SP-9(OP) Symposium on Creep of Concrete
modulus of elasticity for prestressing bars, strands, or wires SP-27(OP) Designing for Effects of Creep, Shrinkage, and
should be determined by tests if not supplied by the manu- Temperature in Concrete Structures
facturer. When that information is not available, the SP-29(OP) Lightweight Concrete
following values may be used: SP-76 Designing for Creep and Shrinkage in Concrete
Bars: 28,000,000 psi (193,000 MPa) Structures
Strands: 27,000,000 psi (186,000 MPa)
Wires: 29,000,000 psi (200,000 MPa) 3.4.2 AREA Manual for Railway Engineering
Chapter 8—Concrete Structures and Foundations, 1985.
3.4—Standard specifications and practices
The standard specifications for materials and practices 3.4.3 ASTM standards
referred to either directly or indirectly in this chapter are A 36 Specification for Structural Steel
listed here. These specifications are constantly reviewed and A 48 Specification for Gray Iron Castings
revised by members of the respective technical committees. A 53 Specification for Pipe, Steel, Black and Hot-
It is recommended that documents with the most recent dates dipped, Zinc-Coated Welded and Seamless
of issue be reviewed in conjunction with previous standard A 82 Specification for Cold-Drawn Steel Wire for
specifications; the latest revisions are not always in conform- Concrete Reinforcement
ance with the requirements of the various agencies and A 123 Specification for Zinc (Hot-Galvanized) Coat-
owners. A list of ACI guidelines and standard practices is ings on Products Fabricated from Rolled,
given for reference. An ASTM-AASHTO specification Pressed, and Forged Steel Shapes, Plates, Bars
cross-reference is given at the end of this chapter for ease of and Strip
343R-34 MANUAL OF CONCRETE PRACTICE

A 153 Specification for Zinc Coating (Hot-Dip) on B 100 Specification for Rolled Copper-Alloy Bearing
Iron and Steel Hardware and Expansion Plates and Sheets for Bridge and
A 167 Specification for Stainless and Heat-Resisting Other Structural Uses
Chromium-Nickel Steel Plate, Sheet, and Strip B 152 Specification for Copper, Sheet, Strip, Plate,
A 184 Specification for Fabricated Deformed Steel and Rolled Bar
Bar Mats for Concrete Reinforcement B 633 Specification for Electrodeposited Coatings of
A 185 Specification for Welded Steel Wire Fabric for Zinc on Iron and Steel
Concrete Reinforcement C 33 Specification for Concrete Aggregates
A 242 Specification for High-Strength Low-Alloy C 109 Test Method for Compressive Strength of
Structural Steel Hydraulic Cement Mortars (Using 2-in. or 50-
A 276 Specification for Stainless and Heat-Resisting mm Cube Specimens)
Steel Bars and Shapes C 150 Specification for Portland Cement
A 416 Specification for Uncoated Seven-Wire Stress- C 171 Specification for Sheet Materials for Curing
Relieved Steel Strand for Prestressed Concrete Concrete
A 421 Specification for Uncoated Stress-Relieved C 191 Test Method for Time of Setting of Hydraulic
Wire for Prestressed Concrete Cement by Vicat Needle
A 441 Specification for High-Strength Low-Alloy C 260 Specification for Air-Entraining Admixtures
Structural Manganese Vanadium Steel for Concrete
A 446 Specification for Steel Sheet, Zinc-Coated C 309 Specification for Liquid Membrane-Forming
(Galvanized) by the Hot-Dip Process, Struc- Compounds for Curing Concrete
tural (Physical) Quality C 330 Specification for Lightweight Aggregates for
A 496 Specification for Steel Wire, Deformed, for Structural Concrete
Concrete Reinforcement C 403 Test Method for Time of Setting of Concrete
A 497 Specification for Welded Deformed Steel Wire Mixtures by Penetration Resistance
Fabric for Concrete Reinforcement C 494 Specification for Chemical Admixtures for
A 500 Specification for Cold-Formed Welded and Concrete
Seamless Carbon Steel Structural Tubing in
C 595 Specification for Blended Hydraulic Cements
Rounds and Shapes
C 618 Specification for Fly Ash and Raw or Calcined
A 501 Specification for Hot-Formed Welded and
Natural Pozzolan for Use as a Mineral Admix-
Seamless Carbon Steel Structural Tubing
ture in Portland Cement Concrete
A 570 Specification for Hot-Rolled Carbon Steel
C 845 Specification for Expansive Hydraulic Cement
Sheet and Strip, Structural Quality
D 98 Specification for Calcium Chloride
A 572 Specification for High-Strength Low-Alloy
Columbium-Vanadium Steels of Structural D 395 Test Methods for Rubber Property—Compres-
Quality sion Set
A 588 Specification for High-Strength Low-Alloy D 412 Test Methods for Rubber Properties in Tension
Structural Steel with 50 ksi (345 MPa) D 413 Test Methods for Rubber Property-Adhesion to
Minimum Yield Point to 4 in. Thick Flexible Substrate
A 615 Specification for Deformed and Plain Billet- D 429 Test Methods for Rubber Property—Adhesion
Steel Bars for Concrete Reinforcement to Rigid Substrates
A 615M Specification for Deformed and Plain Billet- D 448 Specification for Standard Sizes of Coarse
Steel Bars for Concrete Reinforcement Aggregate for Highway Construction
(Metric) D 496 Specification for Chip Soap
A 616 Specification for Rail-Steel Deformed and D 572 Test Method for Rubber Deterioration by Heat
Plain Bars for Concrete Reinforcement and Oxygen
A 617 Specification for Axle-Steel Deformed and D 573 Test Method for Rubber Deterioration in an Air
Plain Bars for Concrete Reinforcement Oven
A 706 Specification for Low-Alloy Deformed Bars D 746 Test Method for Brittleness Temperature of
for Concrete Reinforcement Plastics and Elastomers by Impact
A 709 Specification for Structural Steel for Bridges D 747 Test Method for Stiffness of Plastics by Means
A 722 Specification for Uncoated High-Strength of a Cantilever Beam
Steel Bar for Prestressing Concrete D 792 Test Methods for Specific Gravity and Density
A 767 Specification for Zinc-Coated (Galvanized) of Plastics by Displacement
Bars for Concrete Reinforcement D 994 Specification for Preformed Expansion Joint
A 775 Specification for Epoxy-Coated Reinforcing Filler for Concrete (Bituminous Type)
Steel Bars D 1149 Test Method for Rubber Deterioration—
B 22 Specification for Bronze Castings for Bridges Surface Ozone Cracking in a Chamber (Flat
and Turntables Specimens)
 BRIDGE ANALYSIS AND DESIGN 343R-35

D 1190 Specification for Concrete Joint Sealer, Hot- M 171 Sheet Materials for Curing Concrete
Poured Elastic Type M 173 Concrete Joint Sealer, Hot-Poured Elastic
D 1457 Specification for PTFE Molding and Extrusion Type
Materials M 182 Burlap Cloth Made from Jute or Kenaf
D 1751 Specification for Preformed Expansion Joint M 183 Structural Steel
Fillers for Concrete Paving and Structural M 194 Chemical Admixtures for Concrete
Construction (Nonextruding and Resilient M 195 Lightweight Aggregates for Structural
Bituminous Types) Concrete
D 1752 Specification for Preformed Sponge Rubber M 203 Uncoated Seven-Wire Stress-Relieved Strand
and Cork Expansion Joint Fillers for Concrete for Prestressed Concrete
Paving and Structural Construction M 204 Uncoated Stress-Relieved Wire for Prestressed
D 1850 Specification for Concrete Joint Sealer, Cold- Concrete
Application Type M 213 Preformed Expansion Joint Fillers for Concrete
D 2240 Test Method for Rubber Property—Durometer Paving and Structural Construction
Hardness M 220 Preformed Elastomeric Compression Joint
D 2628 Specification for Preformed Polychloroprene Seals for Concrete
Elastomeric Joint Seals for Concrete Pave- M 221 Welded Deformed Steel Wire Fabric for
ments Concrete Reinforcement
E 328 Recommended Practice for Stress-Relaxation M 222 High-Strength Low-Alloy Structural Steel with
Tests for Materials and Structures 50,000 psi Minimum-Yield Point to 4-in.
Thick
3.4.4 AASHTO materials specifications M 223 High-Strength Low-Alloy Columbium-Vana-
M6 Fine Aggregate for Portland Cement Concrete dium Steels of Structural Quality
M 31 Deformed and Plain Billet-Steel Bars for M 225 Deformed Steel Wire for Concrete Reinforce-
Concrete Reinforcement ment
M 32 Cold-Drawn Steel Wire for Concrete M 230 Extruded Insulation Board (Polystyrene)
M 33 Preformed Expansion Joint Filler for Concrete M 232 Zinc Coating (Hot-Dip) on Iron and Steel
(Bituminous Type) Hardware
M 42 Rail-Steel Deformed and Plain Bars for M 240 Blended Hydraulic Cements
Concrete Reinforcement M 251 Laminated Elastomeric Bridge Bearings
M 43 Standard Sizes of Coarse Aggregate for M 275 Uncoated High-Strength Steel Bar for
Highway Construction Prestressing Concrete
M 45 Aggregate for Masonry Mortar T 26 Quality of Water to be Used in Concrete
M 53 Axle-Steel Deformed and Plain Bars for
Concrete Reinforcement 3.4.5 ASTM-AASHTO specification cross-reference
M 54 Fabricated Steel Bar or Rod Mats for Concrete ASTM specification AASHTO specification
Reinforcement A 36 M 183
M 55 Welded Steel Wire Fabric for Concrete Rein- A 48 M 105
forcement A 82 M 32
M 80 Coarse Aggregate for Portland Cement A 123 M 111
Concrete A 153 M 232
M 85 Portland Cement A 184 M 54
M 105 Gray Iron Castings A 185 M 55
M 107 Bronze Castings for Bridges and Turntables A 416 M 203
M 108 Rolled Copper-Alloy Bearing and Expansion A 421 M 204
Plates and Sheets for Bridge and Other Indus- A 496 M 225
trial Uses A 497 M 221
M 111 Zinc (Hot-Galvanized) Coatings on Products A 572 M 223
Fabricated from Rolled, Pressed, and Forged A 588 M 222
Steel Shapes, Plates, Bars, and Strip A 615 M 31
M 138 Copper Sheet, Strip, Plate, and Rolled Bar A 615M M 31M
M 144 Calcium Chloride A 616 M 42
M 148 Liquid Membrane-Forming Compounds for A 617 M 53
Curing Concrete A 709 M 270
M 153 Preformed Sponge Rubber and Cork Expan- A 722 M 275
sion Joint Fillers for Concrete Paving and B 22 M 107
Structural Construction B 100 M 108
M 154 Air-Entraining Admixtures for Concrete B 15 M 138
343R-36 MANUAL OF CONCRETE PRACTICE

C 150 M 85 RECOMMENDED REFERENCES

C 171 M 171 The documents of the various standards-producing organi-
C 260 M 154 zations referred to in this report are listed below with their
serial designation, including year of adoption or revision.
C 309 M 148
The documents listed were the latest effort at the time this
C 33 M 194
report was written. Since some of these documents are
C 494 M 195 revised frequently, generally in minor detail only, the user of
C 595 M 240 this report should check directly with the sponsoring group
C 618 M 295 if it is desired to refer to the latest revision.
D 98 M 144
D 448 M 43 American Association of State Highway and Transportation
D 994 M 33 Officials
Standard Specifications for Highway Bridges, Thirteenth
D 1190 M 173
Edition, 1983.
D 1751 M 213
D 1752 M 153 American Welding Society
D 2628 M 220 D1.4-79 Structural Welding Code Reinforcing Steel
 BRIDGE ANALYSIS AND DESIGN
Complex construction site (photo courtesy of Ministry of Transportation and
Communications, Ontario)
CHAPTER 4—CONSTRUCTION CONSIDERATIONS
4.1—Introduction
Every project progresses through various stages of devel-
opment, starting with the recognition of the need and ending
with a completed structure. The major steps in the evolution
of a structure are:
a. Preliminary engineering
b. Design
c. Construction
It is obvious that each phase of development affects the
activities within the other phases. This chapter explores the
influence of construction methods and procedures upon the
design process.
4.1.1 Definition—Construction considerations may be
defined as those details, procedures, and construction
sequences that should be incorporated into design to cope with
construction restrictions and limitations while insuring the
constructibility of a structure. For quality assurance systems
ACI 121R provides guidelines in establishing a project
program for describing an organization’s policies, practices,
and procedures to comply with the contract documents.
4.1.2 Examples4-1—Failure to consider construction
restrictions during design generally results in additional
expense and loss of time in rectifying the situation during
construction, or unexpected maintenance expenses after the
bridge is in service.
4.1.2.1 Section size—Some obvious construction
considerations are involved in the design of a concrete
343R-37
section. The section should be large enough not only to carry
the imposed loads, but to physically accommodate the
required number of reinforcing bars or prestressing tendons,
while providing proper clearances and concrete cover. Thus,
the designer should consider the maximum size of aggregate
and the adequate size of a vibrator to be used by the contractor
for proper placement and consolidation of concrete.
If a poor choice of section size is made, a more exacting
concrete placement, a revision of maximum aggregate size,
or a design of the section may become necessary. A change
order may be required with consequent delay of the project
and probable cost increase.
4.1.2.2 Camber—More complex construction consider-
ations are involved in the determination of camber in struc-
tures constructed in stages (Fig. 4.1.2.2). Several factors
have to be assessed quantitatively:
a. Variation of modulus of elasticity of concrete with time
b. Variation in creep rates depending on age of concrete at
the time of loading
c. Variation in creep rates for downward and upward
deflections.
d. Changing loading conditions
e. Effects of temperature variations
If improper camber calculations are made, the appearance
and riding quality of the deck will be affected. The omission
will become obvious some time after the completion of the
343R-38 ACI COMMITTEE REPORT
Fig. 4.1.2.2—Example of staged construction.
structure. The defect, although annoying, will seldom be
serious. Since corrective measures are expensive, they will
rarely be taken.
4.1.2.3 Construction sequence—The construction
sequence is a very important consideration for large, contin-
uous, multispan structures. A typical example is a composite
precast, prestressed I-girder bridge with cast-in-place
concrete deck. If the dimensions of the structure make it
impossible to place the whole deck in one working day, it
becomes necessary to provide a concrete placement diagram,
specifying sequence and size of individual deck placement
sections. The factors to consider are:
a. Maximum size of a reasonable deck placement section
b. Stresses at critical points during all stages
c. Time interval between various sections
d. Strength of concrete to be attained in one section before
another one is allowed
e. Effect of staged construction on deflections
If construction sequencing is neglected, the results can be
serious. During placement of some of the deck sections, the
stresses at critical points in the partially completed girder
may exceed the design stresses, which are based on the
assumption of continuity and composite action of the girder
and the slab.
4.2—Restrictions
The construction phase must be considered during design
to insure that the construction can be economically done in
accordance with the design assumptions and to comply with
construction restrictions imposed on the project by existing
conditions or approving agencies.
Construction considerations originating in design assump-
tions usually deal with:
a. Construction tolerances
b. Stresses and deflections induced by construction sequences
c. Effects of construction joint locations
d. Measures to insure economy of construction
The construction considerations, having their source in
restrictions imposed on the project, are numerous and may
deal with any facet of the structure. The following tabulation
gives examples of typical construction restrictions and their
sources:
Owner—Project schedule (accelerated schedule will influ-
ence selection of the type of structure and construction
sequence).
Approving agencies—Construction clearances (these are
sometimes more restrictive than clearances for the
completed structure, thus influencing layout and type of
structure selection).
Timing of specific construction activities—Encroach-
ments on the flood plain, construction in forests during fire
season, pile driving or other noisy activities next to hospitals
and schools.
Access restrictions—Parks, city streets.
Maintenance of traffic—More and more structures are
being built as replacements or widenings.
Site characteristics—Accessibility (narrow, twisting roads
preclude use of long, prefabricated girders).
Adjacent structures—There may be weight and size limi-
tation.
Climate—Short construction season will influence type of
structure selection.
Materials—Quality of local aggregates may determine
maximum strength of concrete that can be readily produced
(lightweight aggregates, freeze-and-thaw-resistant aggre-
gates are not available at many locations).
Project needs—Traffic requirements may require staged
construction (traffic detours).
Avoidance of possible claims—Blasting, dewatering, pile
driving (in many cases proper design can minimize or even
eliminate these problems).
4.3—Goals
The designer should strive to achieve four primary goals in
the completed structure: sufficient capacity, dependable
durability, economy of construction, and pleasing appear-
ance. In general, sufficient capacity in structures is achieved
by adhering to appropriate codes, although it should be kept
in mind that most codes are based on an acceptable minimum
capacity while quite often a greater capacity may be
warranted or desirable. The achievement of any of the goals
requires engineering judgment.
The goals are interrelated; effort expended to better one
goal invariably changes the others, but not necessarily for the
worse. The goals are also influenced by availability of mate-
rials and by construction methods.
The ideal of constructing a bridge with an excess of capacity
that is maintenance free and has a very pleasing appearance, at
much less than the budget figure, is seldom attainable. Various
factors have to be considered to determine the effect of each
 BRIDGE ANALYSIS AND DESIGN
construction consideration on each goal. Trade-offs can then be
weighed to achieve the optimum result.
The importance of each of the goals varies with the polit-
ical and social environment, as well as the geographical loca-
tion of the project. Adequate capacity must always be
insured. The appearance of a bridge is usually a more sensi-
tive issue in an urban area location than in an industrial or a
rural area (although obviously each case has to be considered
and decided on its own merits).
When cost of construction is of prime importance, construc-
tion methods tend to control design. The designer should
consider several methods to select the most economical way to
build a particular project, recognizing in each method the
applicable restrictions mentioned in the previous section.
Once the most economical method and material has been
determined, the designer decides how to use them to achieve
acceptable appearance and adequate capacity and durability.
New construction methods and material strengths often
create situations in which construction loads may govern the
required strength of part or all of a structure. The designer
should always keep construction loadings in mind. A close
working relationship with qualified contractors should be
developed to insure that allowable stresses are not exceeded
during construction. Contractors are becoming increasingly
involved with engineering requirements. The most efficient
use of labor and materials leads to more engineering-inten-
sive designs that require closer tolerances and higher
stresses. The level of sophistication in design should take
into account a reasonable level of the quality of construction
that can be achieved.
In summary, the design engineer should recognize the
applicable construction considerations during the design
phase; the construction engineer should recognize the impli-
cations of the design criteria and be able to construct the
project within these limitations. The result will be an
economical product that is adequate in capacity, pleasing in
appearance, and durable in service.
4.4—Planning
A conventionally scheduled design project, as opposed to
fast-track projects, is characterized by a sequential progres-
sion in which each activity depends on work performed in a
previous one. Because of this dependence, each phase should
be completed before the next is begun. The design process
should include the following activities:
a. Data collection
1. Traffic during construction
2. Falsework
3. Stage construction requirements
4. Construction restrictions
5. Construction loads
6. Allowable construction stresses
b. Layout
c. Foundations
d. Structure type selection
e. Details
As this list demonstrates, most construction considerations
are based on data collected during the initial phase of design.
343R-39
Any construction requirement or restriction overlooked, during
this period, may be a costly item to consider at a later stage.
Costs of revisions escalate, particularly after various binding
decisions on layout or type of structure have been made.
Thus, it is essential to deal with all construction consider-
ations affected by external constraints at the earliest stages of
project activities. On the other hand, construction consider-
ations originating in design assumptions may be handled
when the affected detail is being designed.
4.5—Site characteristics
In the past “forced solutions,” using standard right angle
structures requiring the realignment of streams or secondary
roads, were often adopted under the guise of economy.
Today it is axiomatic that the structure should fit the site.
Some of the elements dictating the choice of structure
include alignment, length, spans, depth, and foundations.
Lately, with greater emphasis on esthetics, the requirement
of visual harmony with the site in terms of shape, color, and
texture has been added to this list. In addition, some site
characteristics affect the design indirectly by placing various
restrictions on construction.
4.5.1 Site accessibility—Many bridge construction sites
are remote, and accessible only by narrow, winding roads.
Such conditions generally favor structures that require
minimum amounts of materials and field labor. When poor
roads preclude transportation of long prefabricated girders,
girders may be shipped in segments and assembled at the site.
4.5.2 Climate—The various ways climate influences
construction should be considered in design.
4.5.2.1 General—In areas of severe climate, the construc-
tion season may be short, favoring structures that allow a great
deal of shop fabrication. In areas subject to periodic floods or
storms, even the substructure should be designed to permit
some shop precasting and quick field assembly.
Because heating and cooling increase the cost of concrete,
designs with precast in lieu of cast-in-situ elements may be
preferred.
In very cold climates, a warm working area is necessary to
allow the work to proceed. Low productivity and poor
quality of workmanship generally result from work done in
unfavorable conditions.
4.5.2.2 Air-entrained concrete—Air entrainment is used
to develop freeze-thaw durability in concrete. The require-
ments for entrained air and compressive strength should be
specified separately. Mixes with entrained air having
strengths up to 6000 psi (47.5 MPa) can be produced in most
areas. However, the designer should verify the capability of
the local producers.
4.5.3 Materials availability—Materials of adequate
strength and quality may not be available locally.
Concrete strengths of 4000 to 5000 psi (27 to 34 MPa)
using commonly available local aggregates are generally
possible. With superplasticizers and pozzolans, 8000 to
9000 psi (55 to 62 MPa) can be produced with local mate-
rials in most areas. If it is necessary to bring in aggregates
from other sources, the designer should evaluate the benefits
gained by the additional cost.
343R-40 MANUAL OF CONCRETE PRACTICE
High-quality, lightweight concrete aggregates, producing
concrete weighing under 110 lb/ft3 (1760 kg/m3) and having
fc′ = 3500 psi (24.1 MPa), are not readily available at all loca-
tions. The designer should evaluate whether the benefits of
this more expensive, specialized concrete outweighs its
disadvantages.
4.5.4 Temporary foundations—The load-carrying capacity
of the surface soils at the construction site should be consid-
ered. If the falsework has to be supported by driven piles, the
cost of structures constructed on ground-supported falsework
increases and precast options may be more economical.
Another consideration is the ability of the soils to support
construction equipment needed in common construction
operations: pile drivers, cranes, trucks, etc. In swampy or
steep, rocky areas, expensive construction roads may have to
be built. In this case an evaluation should be made of
measures which could reduce pile-driving activities, crane
use, and similar operations.
4.6—Environmental restrictions
Obtaining environmental clearance for a bridge project can
be a complicated process affecting many facets of design and
construction. This section deals only with those construction
restrictions which might affect the design of a bridge.
4.6.1 Falsework—When the use of ground-supported
falsework is prohibited, the restriction should be known
early in the design process. It can then be handled routinely
by selecting an appropriate type of structure. Troublesome
cases arise when the restriction on use of falsework is not
readily apparent in the design plans and is only discovered
during the construction stage.
Use of falsework within a flood plain may be permitted only
between specific dates, resulting in an artificially short
construction season. For all practical purposes this may require
a structure that does not need ground-supported falsework.
The requirements for falsework located next to railway
tracks may be so restrictive as to make the use of such false-
work very expensive and perhaps impractical. For
example, steel posts set in holes filled with concrete and
concrete crash walls may be required. As a result, a struc-
ture not requiring ground-supported falsework may be
preferred for such locations.
4.6.2 Earthwork—Excavation, pile driving, and similar
activities are usually prohibited during the spawning season,
within streams and lakes used by game fish. The restriction
generally lasts only a few months each year and is not a
serious handicap for an average project. However, it should
be considered in the context of the other constraints, so that
the combined restrictions do not make the completion of the
project impossible.
With current emphasis on preservation of environment and
historical heritage, a restriction on construction of access
roads is included in many contracts. In some cases, the restric-
tions are absolute, requiring the use of a cableway or a heli-
copter for construction of the intermediate supports and parts
of the superstructure. Use of precast segments for both the
superstructure and the piers is almost mandatory in such cases.
For construction within areas such as recreational parks,
some agencies require immediate removal and disposal of
excavated material. Such a requirement obviously favors
designs that minimize the foundation and earthwork needed
in construction.
4.6.3 Construction—In many western states, fires during
dry summer months are greatly restricted. On-site field
welding is also prohibited during particularly dangerous forest
fire periods. Such restrictions should be considered during the
design phase when the structure type is selected, particularly
if a steel structure seems to be an attractive alternative.
On construction projects close to sensitive areas, such as
schools and hospitals, there are strict limitations as to the
time of construction activity and amount of noise. The
activity usually involved in such situations is pile driving.
The remedy may be to use auger-cast piles or spread foot-
ings. The former may be very expensive, and the latter may
lead to future problems of support settlements. On the other
hand, a careful selection of working hours may alleviate the
problem. Obviously, the problem is much more easily dealt
with if it is considered during the design phase.
4.7—Maintenance of traffic
Traffic considerations are involved in construction of most
bridges, though there are cases where traffic is entirely
diverted from the construction site. The effects of traffic on
construction will be particularly felt in construction clear-
ances at the crossing. Traffic clearances for construction are
generally smaller than those for permanent structures
because of lower speed limits and the temporary nature of
the clearances. In many instances, however, additional
requirements imposed by the permitting agency may make
the lesser construction clearances unattainable, thus
affecting the layout and span lengths.
4.7.1 Railroad clearances—As an example of the process
involved in determining the governing clearances, consider
the clearance diagram shown in Fig. 4.7.1. The required final
clearance from the center line of track to the face of a pier
may be as small as 10 ft (3 m) for tangent tracks, but the
closest the excavation for the foundation of the pier can be is
8 ft-6 in. (2.6 m), excluding the shoring and bracing. Thus
the final clearance does not control the location of the pier
and the designer should consider the size of footing, the
depth of its location, the likely method of its construction,
and the possibility of obtaining approval for a discretionary
minimum clearance requirement.
4.7.2 Highway clearances—Prescribed temporary
construction clearances often govern layout of spans. A
typical example is the required vertical clearance over free-
ways in California, as shown in Table 4.7.2. The usual
requirement is a clearance of 16 ft-6 in. (5 m) over the trav-
eled way, but the temporary construction clearance may be
as low as 14 ft-6 in. (4.4 m). However, for a structure
constructed on ground-supported falsework, where a 40-ft
(12.2-m) wide opening for traffic is needed, an adequate
depth of falsework may be 2 ft-6 in. (0.75 m) to 3 ft-0 in. (1.0
m). This results in a final clearance of 17 ft-0 in. (5.2 m) to
17 ft-6 in. (5.3 m).
 BRIDGE ANALYSIS AND DESIGN 343R-41

Fig. 4.7.1—Example of typical railroad clearance.4-2

Table 4.7.2—Example of falsework depth and span relationship4-2

Minimum width of traffic Required falsework span Minimum depth required
Facility to be spanned opening, ft (m) Opening width provides for (b), ft (m) for falsework, ft (m)
25 (7.6) 1 lane + (d) 32 (9.8) 1 ft 9 in. (0.53)
37 (11.3) 2 lanes + (d) 44 (13.4) 2 ft 2 in. (0.66)
Freeway
49 (14.9) 3 lanes + (d) 56 (17.1) 2 ft 8 in. (0.81)
61 (18.6) 4 lanes + (d) 68 (20.7) 3 ft 3 in. (1.0)
20 (6.1) 1 lane + (e) 27 (8.2) 1 ft 9 in. (0.53)
32 (9.8) 2 lanes + (e) 39 (11.9) 1 ft 11 in. (0.58)
Nonfreeway 40 (12.2) 2 lanes + (f) 47 (14.3) 2 ft 4 in. (0.71)
53 (15.8) 3 lanes + (f) 59 (17.9) 2 ft 9 in. (0.84)
64 (19.5) 4 lanes + (f) 71 (21.6) 3 ft 5 in. (1.04)
Special (a) 20 (6.1) 1 lane + (e) 20 (6.1) (c) 1 ft 7 in. (0.48)
roadways 32 (9.8) 2 lanes + (e) 32 (9.8) (c) 1 ft 9 in. (0.53)
(a) Uses such as fire or utility access or quasi-public road with very light traffic.
(b) Includes 7 ft (2.1 m) for two temporary guardrails.
(c) No temporary guard railing provided.
(d) 8 ft (2.4 m) and 5 ft (1.5 m) shoulders.
(e) 2 ft 4 in. (1.2 m) shoulders.
(f) 2 ft 8 in. (2.4 m) shoulders.

4.8—Project needs 4.8.1 Construction sequence—The requirement of

Criteria governing the completion of an entire project may
impose restrictive construction considerations on the design
of an individual structure within the project. These consider-
ations may involve a prescribed construction sequence
(perhaps even an accelerated schedule), special construction
loads, or uniformity of details and appearance.
allowing the passage of traffic through the construction site
generally leads to a stage-constructed structure.
4.8.1.1 Partial-width construction—Bridge replacement
projects usually require that the existing structure remain in
service until a new one is completed, or that the new structure
be constructed in the approximate location of the old one.
343R-42 MANUAL OF CONCRETE PRACTICE
In such cases, about one-half of the new structure may be
constructed adjacent to the old structure. After traffic is
shifted to the completed half of the new structure, the old one
is removed and the second half of the new structure is
constructed. Typical construction considerations in such
cases are:
a. The partial-width structure should be stable at all times,
either of itself or through use of temporary bracing
b. The substructure of the new bridge should be of such a
configuration as to permit half-width construction
c. The deformations and creep shrinkage strains of the
structure should be determined and provided for in the
design and construction
d. The part of the structure constructed first should be wide
enough to accommodate the expected traffic
e. If a portion of the existing structure is to be removed
prior to the construction of the new structure, the
amount of removal and staging of the removal should be
determined by the design engineer to insure its stability
f. The protection of the new structure during the removal
of the old structure
4.8.1.2 Partial-length construction—In cases of multi-
span bridges or multilevel interchanges, it may be necessary
to construct the structure in stages, using either intermediate
hinges or construction joints. A typical layout of a crossing
over a highway, as shown in Fig. 4.1.2.2, may involve a span
and a cantilever extending to some point into the adjacent
span. These are constructed first, then the falsework is
removed and the traffic shifted into the opening under the
completed span. The remainder of the structure is then
completed without interfering with rerouted traffic.
Typical construction considerations for such cases are:
a. The initially constructed span may have to be designed
as a simple span. It may be necessary to load the canti-
lever with temporary loads until the full dead load reac-
tion is applied
b. The camber for all affected spans should be calculated,
based on the assumption of a reasonable construction
sequence, taking into account creep and deflections due
to all temporary loads. Improper camber may result in
rough riding decks, poor drainage, and possibly a traffic
hazard.
4.8.1.3 Detour bridges—In many cases, it may be
possible to divert the traffic around the construction site by
means of a detour. Sometimes an existing bridge at another
location can be used. At other times it is necessary to
construct a temporary bridge. Most agencies will impose the
same design standards for temporary bridges as for perma-
nent ones. However, the possibility should be thoroughly
investigated, as substantial savings might be achieved by
minor changes in the design requirements. Typically, the
following construction considerations apply:
a. Used materials may be permitted in the construction of
a temporary bridge
b. The contractor should be given complete latitude in
selecting the basic material for the bridge (timber, steel,
concrete)
c. Use of prefabricated proprietary modular units should
be allowed
4.8.2 Construction loads—For most structures, construc-
tion loads do not represent a major part of design loads. For
structures constructed on falsework, all loads applied during
construction are carried by the falsework. However, there are
cases where construction loads represent major design loads
and thus should be fully considered in the design.
4.8.2.1 Composite and segmental structures—Portions
of these structures support all or part of their own formwork
and construction loads. Also included are structures using a
truss to construct some load-carrying elements that will later
support the formwork and construction loads. Common
construction considerations for these structures are the
construction loads, their magnitude, and location. The struc-
ture may be designed to carry all the construction loads, or
the contractor may be required to modify the original design
for this purpose. Construction loads for structures using
precast girders are usually smaller than the live loads and do
not govern the design. Construction loads for segmental
structures are quite large and generally control the design at
various sections along the girder and at various times during
erection. Requiring the contractor to redesign the structure
for construction loads involves extensive engineering and
expense. Furthermore, such contracts, being difficult to
administer, can lead to disputes and litigation among the
parties concerned, namely, the engineer, the owner, and the
contractor. To avoid these possible difficulties, the designer
may elect to estimate and design for the most likely construc-
tion loads. A conservative assumption may lead to some
overdesign, but can avoid problems in the long run.
4.8.2.2 Earthmoving vehicles—On big freeway projects
it may be necessary to move large quantities of embankment
material from one side of the freeway to the other. Conven-
tionally, one of the many overcrossing structures within the
project is constructed first and used for moving the earth
material by standard capacity dump trucks. However, an
alternate approach used in some cases has proved to be fast
and economical. One structure within the project may be
designed for heavy earthmoving vehicles, with subsequent
saving in cost of material moving. Fig. 4.8.2.2 shows typical
axle loads for such vehicles.
The construction considerations for such cases may be:
a. What is the size and weight of the heaviest construction
vehicle likely to be used by the contractor?
b. Is the provision for two-way traffic economically justi-
fiable?
c. Will the savings on the placement of embankment material
outweigh the additional expense of the structure?
The answers to these questions should be evaluated by the
design engineer working in cooperation with the owner.
A similar situation occurs when large amounts of earth
should be moved from one side of a river crossing to the other.
4.8.3 Standardization—A uniform approach in the design
of a number of structures within one larger project often
results in an economy of the total project.
 BRIDGE ANALYSIS AND DESIGN
Generally each structure should be designed to fit a partic-
ular location; however, some economies may be realized in
using the same type of structure for several locations.
Use of the same material in all the project structures is
preferable. When materials of various strengths are avail-
able, only one or as few as possible should be used. The
economics and the logic of specifying several strengths of
concrete for a given project should be investigated and the
decision made on merits of each individual case. To avoid
placing errors, it is also preferable to use only one grade of
reinforcing bars on a project.
4.9—Design of details
Generally, the economy of a structure is enhanced by
striving for simplicity and repetitiveness in details, making
each part of the structure easy to form, reinforce, fill with
concrete, and strip.
4.9.1 Dimensions—Dimensions of elements should be
large enough to accommodate reinforcement and permit
placement and consolidation of concrete at the most
congested point.
4.9.2 Repetitiveness—Forming of odd shapes is expen-
sive. The multiple use of complicated formwork within a
large project will help to reduce cost. Where voids are indi-
cated, it is often advantageous to allow the contractor to
substitute solid sections where structurally permissible.
Giving the contractor a choice in the use of variations of
details that do not materially affect the strength or the
appearance of the structure is an effective way to increase the
overall economy of a project. However, this permission
should be given in the contract documents.
4.9.3 Slipforming—The overall economy of a project
depends to a great degree on allowing the contractor the
greatest possible freedom in selection of construction methods.
Tall piers often can be constructed most economically by slip-
forming. It is difficult and expensive to slipform piers of a vari-
able cross section. However, if the variability is restricted to
one direction only, slipforming may become competitive.
4.9.4 Soffit lines—The top slab of box girders with widely
spaced stems is usually of variable thickness. For the soffit
line, a series of straight lines will function as well as a curve.
Some saving may be achieved by giving the contractor a
choice between using either a broken or a curved soffit line.
4.9.5 Placement of reinforcement—In cases where the
reinforcement can be positioned a certain distance from the
form face by the use of concrete or plastic blocks, there is
usually no problem in maintaining proper cover. However,
in cases where the reinforcement is next to a finished face
(top reinforcement of slabs) or where the reinforcement is
located inside a deep member, consideration should be given
to some positive means of securing these bars in their proper
places. In concrete slabs where workmen commonly walk on
the top mat, a sufficient number of reasonably rigid bar
chairs should be specified or detailed on the plans to insure
that the constructed effective depth of the slab is as assumed
in design. For bars located within deep members (e.g., two-
layered arrangement of longitudinal reinforcement), addi-
tional supporting reinforcement should be detailed to fix
343R-43
Fig. 4.8.2.2—Axle loads of typical earthmoving vehicles.
their position. The cost of this auxiliary reinforcement is
small compared to the cost of repair.
4.9.6 Placement of anchor bolts—In most instances where
bolts are embedded in concrete, the placement demands a
fairly high degree of accuracy. The simplest method of
insuring the required accuracy involves the use of a prefab-
ricated steel template/spacer to which the bolt heads are tack
welded (incidentally, care shall be exercised to avoid welding
on any other part of the bolt). The template is wired to adjacent
reinforcement (if the reinforcing is too far away, additional
reinforcement should be called for) and can be positioned
accurately and kept in place during placement of concrete.
Some agencies require that the anchor bolts be set in holes
drilled after the concrete has hardened. In this case the
designer should space the reinforcement to provide ample
tolerance to clear the drilled hole.
4.9.7 Hinges—The term “hinge” as used in the discussion
that follows denotes a joint located within a structural
member and constructed in such a way as to permit the
assumption of no moment at the joint while allowing transfer
of vertical and/or horizontal shear.
In terms of maintenance problems, hinges have been
found to be particularly troublesome. These problems are
caused by a combination of factors, some originating in
design and others in common construction practices. In the
past, there has been some difficulty in assessing all the forces
acting upon various parts of a common hinge. Most hinges,
even with the best expansion bearings, involve axial tension
in the direction of the girders. ACI 318, Section 11.9, Special
Provisions for Brackets and Corbels, is quite helpful in clar-
ifying this point.
Another point to recognize is that the ever-present “bump”
at the hinge location can become quite large due to settle-
ments, creep deformations, and deterioration. This can cause
343R-44 MANUAL OF CONCRETE PRACTICE
impact forces much greater than those used in design. For
conventionally reinforced structures, the combination of high
bending and shear stresses at most hinges requires placement
of large amounts of horizontal and vertical shear reinforce-
ment in addition to tensile reinforcement. This congestion of
reinforcement makes the placement and consolidation of
concrete difficult, and often results in rock pockets and honey-
combs. Adequate proportioning of sections and proper super-
vision of construction are necessary. In many cases it may be
advantageous to replace some of the conventional mild steel
reinforcement with short post-tensioning tendons. The post-
tensioning systems used should be those that allow stressing
and anchorage of short tendons without seating loss.
A similar instance of theoretical hinges at bottoms (or
tops) of columns needs careful attention to detail. For
initially applied loads (dead loads) and depending on the
care used in construction, these hinges will have almost no
moment. However, after the application of large axial loads,
and after some deterioration of the hinge materials takes
place (e.g., expansion joint material compresses and
becomes hard), the moments being transferred across such
hinges can become quite large. Generally, the stresses
induced at the hinge itself are rarely objectionable; however,
at other joints of the structure, the additional moments and
stresses caused by the partial fixity of the hinges may be
greater than anticipated in the design.
4.9.8 Fixed end supports—In its pure form, this type of
support is seldom used because full fixity cannot be
achieved. The designer needs to review the design assump-
tions very carefully, particularly in view of likely location of
construction joints and number of reinforcement splices. A
100 percent fixity is probably not a justifiable assumption,
and therefore some increase in adjacent positive moments in
design should be provided.
4.9.9 Accuracy of construction—The design assumptions
and design methods do provide some consideration of the
inaccuracy of construction. The phi factor used in load factor
design does take into account some variability of quality of
material, and the column design criteria assume a certain
eccentricity of loads.
The accuracy of construction to be considered here deals
with cases arising out of construction methods and proce-
dures beyond these allowances that need to be considered by
the designer.
4.9.9.1 Pile location—All pile footings should be large
enough to accommodate the piles that may be driven or
drilled out of exact location. The design should provide for a
tolerance of 3 in. (75 mm) out of location for the pile group.
A 3-ft (1.0-m) c. to c. minimum pile spacing and a 1.5-ft
(0.5-m) minimum edge distance are considered normal for
12-in. (300-mm) piles. Where a greater accuracy is needed
(such as single-line pile bents consisting of driven or drilled
piles and pile extensions above original ground), a pile-
locating template should be used.
The designer is cautioned that a pile group having less than
a 3-in. (75-mm) eccentricity may have within it piles located
6 in. (150 mm) or more out of location. The ACI 318, Section
15.5 provision for reduction of shear if the pile center lies
within one-half the pile diameter of the d/2 distance from the
column face should be used with judgment.
4.9.9.2 Precast elements—Precast segments have been
used to assemble I-girders, box girders, slabs, whole super-
structures, piers, columns, abutments, and walls. With use of
match casting, there is a reasonable fit between the indi-
vidual segments. Problems usually occur with longer girders
(or higher piers) where small inaccuracies may add up to a
sizable deviation from anticipated alignment. For girders,
this may produce unsightly vertical sag or lateral wobble;
piers and columns may have a bowed shape or be out of
plumb in any direction.
The task of the designer is to provide for this eventuality
by incorporating some method of correction into the
design. For sagging girders it may be a provision for addi-
tional ducts for corrective post-tensioning, as well as an
occasional cast-in-place segment. Cast-in-place segments
also may be used for correcting alignment of piers and
columns. Thin shims in conjunction with epoxy mortar also
have been used for this purpose.
The design of segmental precast structures or parts of struc-
tures should readily permit adjustments during construction.
4.10—Selection of structure type
When several types of structure seem to be equally
economical and suitable for a particular location, the deci-
sion may be deferred to the bidding process through prepa-
ration of alternative designs. Obviously, the additional
expense of preparing two or more designs has to be justified
by the possible savings. Such procedures are usually reserved
for major structures. The two main requirements for such
alternative designs—the equivalency of strength and appear-
ance—are not easily met unless the intent is accurately
described or illustrated. The additional time needed for review
of such alternative designs also complicates the process.
The following sections outline some of the construction
considerations connected with various types of concrete
structures:
4.10.1 Concrete slab bridges—A concrete slab bridge is
the simplest and least expensive structure that can be built
within the span limitations for this type of superstructure. It
can be conventionally reinforced, pretensioned or post-
tensioned. It can be built on ground-supported falsework or
constructed of precast elements.
4.10.1.1 Cast-in-place bridges—The simplicity of the
concrete slab bridge may lead to overconfidence on the part
of the contractor, often resulting in poor workmanship.
Falsework founded on soft or shifting soils has led to differ-
ential settlements and bulges in the soffit line during
construction. The solution may be to specify allowable soil-
bearing values for falsework design or to require use of pile-
supported falsework.
Another problem often encountered during construction of
slab bridges is the difficulty of keeping the top mat of rein-
forcing bars in its proper position. Since the workmen
usually walk on these bars, supports have to carry construc-
tion loads, as well as the weight of the bars. If the size and
the spacing of the bar chairs are not shown on the plans, only
 BRIDGE ANALYSIS AND DESIGN 343R-45

the minimal amount will be used during construction with

undesirable consequences.
The designer, therefore, should make sure that a sufficient
number of reasonably rigid bar chairs is called for on the
plans to keep the top mat of reinforcement in its place.
4.10.1.2 Precast slab bridges—Precast slab bridges
constructed as simple spans often exhibit poor riding quali-
ties due to the discrepancy between the camber and actual
deflections. Transverse joints at piers further impair the
riding quality of the deck and often present serious mainte-
nance problems. Providing reinforcement in the topping slab
to develop continuity can alleviate this problem.
Precast slab bridges constructed with unreinforced grout
keys in the longitudinal joints, between individual units, also
develop maintenance problems. The grout keys often fail,
allowing differential live load deflections between units. An
asphalt concrete overlay does not solve the problem, since the
cracks usually continue to show up through repeated overlays.
A reinforced concrete key or other positive connection
between individual precast units, such as transverse post-
tensioning, is helpful.
4.10.2 Reinforced concrete T-beams—This type of struc-
ture is usually constructed on ground-supported falsework. It
is most suitable for bridges of short span lengths of 30 to 80 ft
(9 to 24 m).
4.10.2.1 Reinforcement—Due to the narrow stem width,
a multilayer arrangement of reinforcement in the positive
moment region is required. This makes concrete placement Fig. 4.10.2.3—Examples of longitudinal expansion joints
difficult, often resulting in rock pockets. The selected width for wide structures.
should accommodate the reinforcement while allowing easy
placement and consolidation of the concrete. The decision as to which detail to use will depend on the
The problem of maintaining the top mat of reinforcement total width of structure; the magnitude of initial deflection
in its proper position, as discussed in Section 4.10.1.1, is and rotation to be accommodated by the joint; the amount of
even more serious for thin slabs where a shift of reinforce- creep deflection accumulated by the first half before the
ment by 1/2 in. (12 mm) in a 7-in. (178-mm) slab will mean second half is constructed; and finally by the need to provide
a change in effective depth of 10 percent. for maintenance of traffic.
Girder stirrups and bent-up slab reinforcement bars do To minimize confusion of the drivers, the joint should be
provide some support for the top mat; however, the designer located on a lane line or within a median.
needs to be sure that the supports provided will prevent a 4.10.3 Precast, prestressed girders—These are most suit-
downward displacement of the reinforcement under the able for locations where the use of falsework is either prohib-
probable construction loads. ited, impractical, or too expensive. The construction time is
4.10.2.2 Construction joints—Usually the T-beam usually shorter than that needed for cast-in-place girders.
superstructure is constructed in two separate stages: the The girders are designed to carry dead load and construction
stems and the slabs. To minimize cracks in the tops of stems loads as simple span units. Live load and superimposed dead
due to temperature and shrinkage stresses, as well as possible load design may or may not use continuity and composite
differential settlement of falsework, some longitudinal action with the cast-in-place slab.
reinforcement should be placed within the stems just below The considerations mentioned for T-beams also apply to
the construction joint. wide structures constructed with precast prestressed girders.
4.10.2.3 Longitudinal joints—For very wide structures The discussion of possible downward displacement of rein-
constructed full width, concrete placement and finishing forcement during concrete placement mentioned in Section
becomes difficult and expensive. Temperature stresses and 4.10. also applies.
movements imposed on the substructure become rather 4.10.3.1 Transportation and handling—Most state
large. For these reasons, a longitudinal joint within the struc- highway departments require a transportation permit for any
ture becomes almost a necessity for bridges wider than about load over 80 ft (24 m) long, and many will not issue permits
60 to 70 ft (18 to 21 m). for loads over 100 ft (30 m) in length. Thus, long girders may
A typical detail in such cases may involve: a construction have to be brought to the site in segments and assembled
joint, or an expansion joint, or a construction joint and an there, resulting in increased cost of field work. Poor access
expansion joint with a closure slab as shown in Fig. 4.10.2.3. roads can also require segmental construction.
343R-46 MANUAL OF CONCRETE PRACTICE
Long, precast, prestressed girders are heavy and may be
laterally unstable (particularly if the compression flange is
narrow) until incorporated into the structure. Firm ground is
needed to store the girders, as well as to support the required
lifting cranes. Pickup points should be determined before-
hand and lifting hooks installed. Once erected, and until the
diaphragms are cast, the girders should be braced.
4.10.3.2 Camber—Multispan simple span bridges using
precast, prestressed girders do not usually provide a comfort-
able ride. Some of the steps that can be taken to minimize
this problem include:
a. Make girders continuous for live loading by casting
concrete between ends of girders
b. Use a liberal depth for the girders
c. Specify a maximum length of time that the girders may be
stored without being loaded with at least some dead load
4.10.4 Nonprestressed reinforced concrete box girders—
This type of structure is adaptable for use in many locations.
It has been used in both metropolitan and rural areas, for
single structures, or for entire interchanges. The popularity
of this type of structure lies in its economy for a wide range
of spans and layouts and its reasonably good appearance.
4.10.4.1 Reinforcement—The stem thickness selected
should accommodate the reinforcement and allow proper
placement and consolidation of concrete. Particularly in the
case of boxes over 7 ft (2.13 m) deep, a stem thickness larger
than the frequently used minimum of 8 in. (200 mm) should
be considered.
The discussion of possible downward displacement of
reinforcement during concrete placement mentioned in
Section 4.10.2.1 is equally valid here.
4.10.4.2 Construction joints—Temperature and
shrinkage stresses, as well as possible settlement of false-
work after the initial placement of concrete, often result in
vertical cracks in the webs. Additional horizontal reinforce-
ment in the webs near the construction joint has been found
helpful in restricting the size of these cracks.
4.10.5 Post-tensioned concrete box girders—Post-
tensioned box girder construction affords many advantages
in terms of safety, appearance, maintenance, and economy.
Long spans may be constructed economically, thereby
reducing the number of piers and eliminating shoulder obsta-
cles at overpasses. Obstacle elimination greatly enhances the
recovery area for out-of-control vehicles.
4.10.5.1 Falsework—Construction methods for building
post-tensioned concrete box girder bridges are many and
varied. Construction on ground-supported falsework is prev-
alent in the western part of the United States. In recent years
this method has gained acceptance in the eastern and
midwestern states. The use of ground-supported falsework
construction will increase as projects are evaluated to deter-
mine the most economical construction method.
Falsework construction methods may be modified when
site restrictions dictate the elimination of ground-supported
falsework. The falsework may be supported by an auxiliary
truss spanning from pier to pier. An example of this method
of construction is the Denny Creek Bridge (Reference 4-3)
located in the Snoqualmie National Forest east of Seattle,
Washington. The extreme ecological sensitivity of the area
required an esthetically pleasing structure with minimal
damage to the environment. To satisfy the restrictions, the
winning contractor chose to cast the box girder section in
forms, supported by a traveling truss. The truss, 330 ft (100 m)
long and weighing 540 tons (4800 KN), moved forward span
by span as construction progressed. Denny Creek Bridge was
the first instance of the use of this system in the United States.
Another modification of ground-supported falsework
occurred on the Napa River Bridge4-4 in California. False-
work towers at 70-ft (21-m) intervals with wide flange
girders were used to support cast-in-place box girder cantile-
vers. The reuse of the towers and forms had an economic
advantage over conventional construction. In addition, a
required 70-ft (21-m) navigation channel was maintained
during construction.
These are only a few of the possible modifications of false-
work construction; many others have been successfully
employed and new ones will be developed. Possible restric-
tions on each project should be thoroughly investigated so
the most economical method of construction can be used.
4.10.5.2 Construction options—Post-tensioned box
girder construction offers many options to both designers
and builders for a successful project at minimum cost. The
size of components cast at one time may be varied from
partial spans or partial cross sections to entire spans. The
components may also be precast and assembled with the
post-tensioning tendons. The prestressing tendons, whether
pretensioned or post-tensioned, can be located to place the
required forces in the most efficient manner. The tendons
may be straight, draped, continuous throughout the structure,
or of partial lengths. For instance, the contractor on the Napa
River Bridge used a longitudinal loop tendon system. The
12-strand tendons extended from the end of one cantilever to
the other cantilever, where they looped around and returned.
Partial stressing of post-tensioning tendons may control
shrinkage and temperature cracks until the concrete gains
sufficient strength to allow full prestress application. Proper
analysis of the structure and the use of some of these options,
either alone or in combination, will result in cost savings
while providing structural integrity.
4.10.6 Post-tensioned segmental construction4-5,4-6—
Since about 1970, when this European technology was intro-
duced into the United States, segmental construction has
added a new dimension to the design and construction of
post-tensioned concrete box girder bridges. The basic
concept is to provide cost saving through standardization of
details and multiple use of construction equipment.
Segmental concrete bridges may be precast by either the
long line or short line system. The segments may also be
cast-in-place. They may be erected by the method of
balanced cantilevers, by progressive placement span by
span, or by launching the spans from one end. Almost any
combination may be used to provide the most cost-efficient
use of materials, labor, and equipment. Both the designer and
contractor should have the opportunity to evaluate and
choose the proper combination.
 BRIDGE ANALYSIS AND DESIGN
4.10.6.1 Standardization—Standardization is accom-
plished by varying the details as little as possible. The webs
may be sloped or vertical, but preferably should remain a
constant thickness for the entire length of the structure.
Ideally the only variation would be the spacing of the shear
reinforcement.
Most existing projects have incorporated sloped webs.
Many believe sloped webs are esthetically pleasing; as a
practical matter they facilitate form stripping. However,
engineers should consider the possible problems encoun-
tered by sloping the webs of variable depth box girders. A
constant web slope means the width of the bottom slab
decreases as the depth of the section increases. The forms
should be fabricated to accommodate the variable dimensions.
Sometimes the saving in substructure quantities due to reduced
superstructure weight justifies the additional form cost.
The top slab should remain of constant cross section. An
exception occurs when minor grade variations are accom-
plished by increasing or decreasing the top slab thickness.
Any variations in the slab thickness should be confined to the
top of the slab to eliminate modification of the inside forms.
The top slab may be of variable depth in the transverse
direction. The design of the box girder section in the trans-
verse direction is based on rigid frame analysis requiring a
moment-carrying capability at the joints. The thickening of
the joint area where the webs connect to the slab provides the
moment capacity along with extra concrete to carry the shear
stress occurring between the web and slab.4-7
Transverse prestressing of top slabs of 40 ft (12 m) or
greater width, in conjunction with variable depth, proves
especially advantageous from the viewpoint of economics
and durability. If the transverse tendons are located close to
the centerline of the slab at the ends of the cantilevers and
extend in a straight line through the slab, the prestressing
force will be in the correct position, relative to the neutral
axis, to carry both the negative and positive moments. This
of course assumes a normal crown. The same will not be true
for segments with constant superelevation or with superele-
vation transition.
The bottom slab should have a constant thickness in both
directions throughout, with two exceptions. The bottom slab
may have to be thickened in the transverse direction at the
web junctions for the previously mentioned reasons. Also, if
the segments are erected by balanced cantilevers, the bottom
slab generally has to be thickened over the piers to carry the
compressive stresses. However, the slab should be back to
normal thickness within two or three segments away from
the pier to return to normal casting procedures as quickly as
possible and to minimize dead load.
4.10.6.2 Construction loads—The concept of construc-
tion loads having an influence on design as discussed in
Section 4.3 becomes very important in segmental concrete
design and construction. The amount of prestressing
required in some areas may well be governed by construc-
tion loads. In other areas, service loads may govern. For the
design to be complete, all loads should be considered even
though the exact location and magnitude of the construction
loads may not be known.
343R-47
4.10.6.3 Design procedure—The proper design proce-
dure is to investigate methods available to construct the
bridge and to base the design on the most probable and the
most economical method, recognizing the pertinent restric-
tions. Not only the method of construction, but the time or
rate of construction, are important because of the many time-
dependent factors (such as concrete creep and shrinkage)
that can have significant effects on stresses and deforma-
tions. The designer should consider details of the assumed
construction method, including locations and magnitudes of
construction loads when preparing the plans and specifica-
tions. The project may then be opened up to alternative
methods of construction consistent with the restrictions and
with the stipulation that allowable stresses are not to be
exceeded, including secondary and time-dependent effects.
If construction loads are exceeded, or the timing is signifi-
cantly different, it will be necessary for the contractor to
provide a redesign.
In this way the contractor has the opportunity to develop
the most economical method. Three situations can occur:
a. The contractor can agree with the designer's assump-
tions and build the bridge according to the plans and
specifications.
b. The contractor can develop an economical construction
scheme with smaller construction loads and redesign the
bridge to save material and thereby produce some cost
saving.
c. The contractor can develop a more economical
construction scheme with larger construction loads and
redesign the bridge to provide some cost saving.
This system has some minor drawbacks. The projected
cost savings should exceed the redesign costs by a signifi-
cant amount to make the effort worthwhile. Included in the
redesign costs should be the engineering costs involved in
reviewing and approving redesigns. However, engineering
costs are usually low when compared with construction costs
on a project. Contractors have to spend money in preliminary
designs to verify methods without any assurance of getting
the work. Unsuccessful bidders probably will pass this cost
on to subsequent projects.
The real advantage of this system is that it provides the
contractor with an opportunity to bid the method most suit-
able to the contractor's expertise. Therefore, the cost saving
is obtained through closer competition.
4.11—Construction problems4-8,4-9
Prevention of the following examples of construction
problems that have been encountered on concrete bridge
projects should be considered during design.
4.11.1 Cracking due to shrinkage and creep—The subject
of shrinkage and creep is quite complicated and involves a
large number of factors, some easy to calculate and others
difficult to predict. These factors include: strength of
concrete, age at which the concrete will be subjected to
stress, magnitude of maximum stresses, temperature varia-
tions, quality of concrete, quality of curing, and general
quality of construction.
343R-48 MANUAL OF CONCRETE PRACTICE
The designer works with a set of well defined design spec-
ifications and design procedures. The product is expected to
be a properly designed structure that can be constructed and
put into service on time, within budget, and without any
major problems. The problems that develop usually come
from factors that are at variance with the assumptions used
in design. These variations can arise from changes in
weather, variability of construction materials, and nonuni-
formity of workmanship. There are cases where rates of
shrinkage have varied by as much as 100 percent from one
batch of concrete to another on the same project. Similarly,
variations of 50 percent in creep rates are not uncommon.
Therefore, it is prudent and cost-effective for the designer to
consider several rates of creep and shrinkage and to assess
the probability of their occurrence and their effects on the
structure, versus the cost of mitigating measures.
Some measures that have been used to reduce the undesir-
able effects of excessive shrinkage rates are: stage construc-
tion of long structures, selective use of high cement content,
concrete additives, and liberal use of continuous nonpre-
stressed reinforcement.
Another common reason for variations in creep rates is the
discrepancy between assumed and actual concrete strength
at the time of load application. This is particularly true in
case of loads applied at a later time in the age of concrete;
e.g., railings, curbs, overlays.
4.11.2 Temperature cracking—This problem often mani-
fests itself on box girders where a very thick bottom slab is
used to provide the necessary compressive area at the
supports. Heat is generated by the hydration of the cement,
with additional input from hot weather in some instances. The
remedies to be considered during the design stage might be:
a. Reduction of slab thickness (increased depth of struc-
ture)
b. More longitudinal reinforcement within the section
c. Reduction of applied moments by use of temporary
supports, stays and tendons
The designer might also specify:
d. A reduction in rate of concrete placement
e. A reduction in cement content
f. The use of retarders and/or pozzolans in the mix
g. Cooling of mixing water and aggregates
In cases where the outdoor temperature represents a large
part of the heat input, use of insulated forms should be consid-
ered. As always, the provision of required minimum ratios for
temperature reinforcement should not be neglected.
4.11.3 Cracking due to tendons—Cracking of concrete
along planes tangential to tendons has been observed on
several structures. This cracking is the result of stresses
exerted by the tendons on the concrete at locations where
there is change in tendon alignment, whether intentional (as
in curved girders, or in girders with draped tendons, etc.) or
accidental (insufficient supports for tendons or workers step-
ping on them).
In cases of intentional curvature, the forces can be calcu-
lated and reinforcing provided as needed for tension in the
plane normal to the tendons. In cases of unintentional curvature
or kinks in the tendon path, the designer should anticipate
and estimate not only in which direction the cracking may
occur, but should also evaluate the cost of additional rein-
forcement versus the cost of better inspection. The use of rigid
ducts and close spacing of duct supports is quite helpful.
4.11.4 Crushing of ducts—This problem occurs on post-
tensioned structures in areas of duct curvature where ducts
are closely placed or actually bundled. As a result, there is
only a thin layer of concrete (sometimes none) to resist the
radial tendon forces. This problem has surfaced on curved
girders, as well as on girders with tendons draped in the
vertical plane. The following remedies are available to the
designer for consideration in these situations:
a. The designer should be satisfied that the concrete
section is large enough to accommodate the required
reinforcement and ducts. “Large enough” is intended to
mean that there is enough concrete around each duct to
resist and distribute, without distress, the horizontal
and/or vertical shears imposed by the tendons on the
concrete. In practical terms, it means that the designer
should select a comfortable depth for the structure,
sufficient width for the girders, and reasonable spacing
for the diaphragms.
b. In cases of unavoidably tight tendon spacing, the
designer should consider specifying metal saddles or
spacers between the ducts. Metal duct saddles have been
used on some projects and are available commercially.
c. In some cases where it is too late for the other remedies,
a carefully thought out post-tensioning sequence will
help, although duct crushing may still occur. Grouting
of such ducts will be difficult and may result in incom-
pletely grouted ducts.
4.11.5 Construction joints—Almost all structures are
designed on the assumption of a monolithic construction.
However, it is very seldom that a structure can be
constructed in such a manner; most structures have a number
of construction joints and some (such as segmental bridges)
have many of them. In situations where there will be a large
number of construction joints within a structural member,
there is normally no need to specify their location; however,
in cases of only a few construction joints in a span or a pier,
it is best to limit their location. Construction considerations
in connection with construction joints are basically the need
to assure a monolithic behavior of the structure, despite the
presence of some degree of discontinuity represented by the
construction joint. Continuity is achieved by insuring the
transfer of joint forces (axial forces, moments, and shears)
across the joint without distress.
4.11.5.1 Joint location—In cases where only a few joints
are used, the preferred location in the span is at or near an
inflection point. Here the moment is near zero, the shears are
lower, and deflection is minimal—conditions favorable for a
distress-free construction joint.
In cases where many construction joints are used in a span
(segmental construction), the location of the joints depends
on the construction method and available equipment and is
usually not specified. Where joints are exposed, regularity of
spacing and uniformity of appearance make the best impres-
sion and thus should be specified.
 BRIDGE ANALYSIS AND DESIGN
4.11.5.2 Joint configuration—Transfer of axial forces
across the construction joint usually presents no problem as
long as the joint surface is normal to direction of forces and the
joint is always in compression, or there is sufficient reinforce-
ment across the joint to properly carry the tensile forces.
Obviously the size of compressive or tensile forces to be
transferred across a construction joint will depend to a great
degree upon the rates of creep and shrinkage. In many struc-
tures, due to an unexpectedly large shrinkage rate, the
prediction of compression on a given joint has been in error.
Tensile stresses acting on the unreinforced plane have
caused unsightly cracking and distress. Thus, the designer
should check the axial forces on all construction joints,
assuming varying rates of shrinkage and creep, and provide
reinforcement across the joint for all doubtful cases. An
alternate approach (although more conservative and expen-
sive) is to provide continuous reinforcement across all
construction joints as a policy requirement.
Transfer of shear forces across a construction joint
presents a greater challenge and needs careful consideration.
In slabs carrying direct traffic loads, a continuous shear key
consisting of a ledge and cantilever has been used for many
years without major problems, and thus may be considered a
preferred solution for vertical shear transfer.
Vertical shear transfer across a construction joint within a
girder web is a problem for segmental construction. The
basic approach has been to develop the shear key in a form
of mating serrated surfaces. Two configurations have been
used: either a great number of shallow keys, or a small
number of large keys.
Both types have had some successes and some failures. In
regions of high shears, small shear keys are not usually
adequate to transfer and distribute the forces beyond the
construction joint plane. On the other hand, the large keys,
although able to distribute the forces, often fail because of inad-
equate (or missing) reinforcement. Thus the solution seems to
be to use small, unreinforced keys in regions of low shears and
large, well reinforced keys in regions of high shears.
4.11.6 Cracking at anchorages—All research dealing
with stress distribution behind anchor plates indicates the
presence of tensile stresses in the vicinity of anchorages and
the need to reinforce the concrete for these stresses.
Most stressing system manufacturers are aware of this and
will usually require a proven and safe arrangement of a grillage
behind the anchorage.
Where cracking of concrete at the anchorages has
occurred, it was usually the result of:
a. Reduction in the amount of reinforcement
b. Misalignment of tendons and anchorages
c. Poor quality of concrete
The designer may have some control over these items during
shop plan review; however, the main burden of insuring quality
construction rests with the contractor's engineer.
4.11.7 Misalignment of ducts—The usual path of
prestressing ducts called for on the plans is a smooth curve.
Any dips, peaks, or kinks will cause high local stresses,
resulting in cracking of concrete and other undesirable
effects. To prevent this, the contractor should have experi-
343R-49
enced and knowledgeable engineers in charge of construc-
tion, with authority to correct poor construction practices
long before they become “cast in concrete,” difficult and
expensive to repair.
4.11.8 Miscellaneous—Over the years, various problems
with post-tensioned concrete have surfaced, been solved and
forgotten, to reappear elsewhere, and to demand new solu-
tions. Some of these are as follows:
a. Drainage pipes installed at low points on all ducts would
correct this problem.
b. Use of watertight tape and care in making duct splices
should be helpful in this respect.
c. In most cases, this problem has been found to result
from improper alignment between the tendon and the
anchorage. Any deviation from a 90-deg (1.57-rad.)
angle between tendon and anchorage will result in large
bending moments on the anchorage plates and on the
tendons, leading to overstress and cracking. In some
cases, the misalignment occurred after stressing because
of rock pockets behind the anchorages. In all cases, the
quality of construction was at fault, which is something
that the designer cannot control with a great degree of
certainty.
4.12—Alternate designs
4.12.1 General—Experience indicates that the lowest bids
are received on those projects that allow the greatest flexi-
bility in use of available construction methods and materials.
Thus, during design it is imperative to avoid introducing
unnecessary restrictions upon the anticipated construction
process. Structures designed with only one construction
method or prestressing system in mind rarely turn out to be
economical. Naturally, there are exceptions, specifically in
situations where the site conditions or the approving agen-
cies impose restrictions upon the designer and the contractor
in selection of type of structure or the method of construc-
tion. However, even under these conditions, permitting the
contractor some latitude within defined limits, will result in
a more economical project.
4.12.2 Value engineering change proposals4-10—Value
engineering has become one of the accepted tools for
reducing construction costs and for introducing innovative
ideas and methods into the construction industry. However,
a questioning attitude still persists among some engineers
toward value engineering change proposals. This attitude is
founded in part upon the mistaken belief that accepting a
change is an admission of a flaw in the original design. The
main point to keep in mind is that the initial designer and the
contractor are designing and constructing two different
structures. The designer had to produce plans for a structure
that could be built economically by the largest possible
number of contractors, whereas the contractors would rather
build the structure in a way most suitable to their expertise,
their equipment, and often their accounting methods. Thus,
acceptance of a value engineering change proposal should
not be considered an acknowledgement of a flaw in design,
but rather another construction consideration that came to
light too late to be considered in original design.
343R-50 MANUAL OF CONCRETE PRACTICE
The drawback to a value engineering change proposal is
that it essentially forces the contractor to bid the project
according to the designer's plans, with the hope of negoti-
ating for the more economical scheme. During the negotia-
tions there is always the possibility of the contractor's not
reporting the entire cost savings. The contractor cannot base
the bid on a more economical scheme since there is no assur-
ance that the value engineering proposal will be accepted.
4.13—Conclusions
On most bridge projects, the designer, after reviewing the
design considerations given in this chapter, can arrive at a
proper decision or decisions. For projects at unusual sites or
of unusual shapes or of major size, the designer should seek
the advice of an engineer experienced in construction, partic-
ularly in the type of construction being considered. If such an
engineer is not on staff, the designer should consider hiring
a subconsultant who is experienced or should confer with
one or more local contractors who are generally willing to
share their knowledge and expertise. If these are not readily
available, the designer should secure the services of a
construction consultant having the necessary expertise.
RECOMMENDED REFERENCES
The documents of the various standards-producing organi-
zations referred to in this report are listed here with their
serial designation, including year of adoption or revision.
The documents listed were the latest effort at the time this
report was written. Since some of these documents are
revised frequently, generally in minor detail only, the user of
this report should check directly with the sponsoring group
if it is desired to refer to the latest revision.
American Concrete Institute
121R-85 Quality Assurance Systems for Concrete
Construction
318-83 Building Code Requirements for Reinforced
Concrete
CITED REFERENCES
4-1. Tadros, G.; Ghali, A.; and Dilger, W.; “Construction
Stages and Design Considerations for a Curved Box-Girder
Bridge,” Canadian Journal of Civil Engineering, V. 9,
No. 3, 1982.
4-2. “Highway Design Manual,” Department of Transpor-
tation, State of California (p. 18 and p. 20).
4-3. “Post-Tensioned Box Girder Bridge Manual,” Post-
Tensioning Institute, Phoenix, Ariz., 1978.
4-4. Podolny, Jr., Walter, “Segmental Bridge Construction
in the U.S.—A Look Ahead,” Long Span Concrete Bridge
Conference, Hartford, Conn., Mar. 1980.
4-5. Joint PCI-PTI Committee on Segmental Construc-
tion, “Recommended Practice for Precast Post-Tensioned
Segmental Construction,” PCI Journal, V. 27, No. 1, Jan.-
Feb. 1982.
4-6. Harwood, Allan C., “I-205 Columbia River Bridge—
Design and Construction Highlights,” PCI Journal, V. 27,
No. 2, Mar.-Apr. 1982.
4-7. “Precast Segmental Box Girder Bridge Manual,”
Post-Tensioning Institute, Phoenix, Arizona, Prestressed
Concrete Institute, Chicago, Ill., 1978.
4-8. Gerwick, Jr., Ben C., “Causes and Prevention of Prob-
lems in Large-Scale Prestressed Concrete Construction,”
PCI Journal, V. 27, No. 3, May-June 1982.
4-9. Freyermuth, Clifford L., “Post-Tensioning Details
for Long-Span Bridges,” PCI Journal, V. 27, No. 6, Nov.-
Dec. 1982.
4-10. “Value Engineering in Preconstruction and
Construction,” 1981, Transportation Research Board
NCHRP SYN 78.
 BRIDGE ANALYSIS AND DESIGN 343R-51

CHAPTER 5—LOADS AND LOAD COMBINATIONS

5.1—Introduction procedure should be done by the contractor. The stability of

The loads recommended herein reflect current bridge precast members, during and after construction, should be
design criteria set forth in AASHTO, the Ontario Highway investigated and provisions made as needed. Effects of
Bridge Design Code,5-1 and recent studies5-2 for a wide member shortening and redistribution of loads during
variety of concrete bridges. In addition, the organization prestressing should be considered.
sponsoring the project may specify loads and criteria which Environmental loads due to wind, earthquake, and thermal
will govern the design. The designer should look for and effects should be considered during construction using an
consider any unique or unusual conditions that may exist at appropriate return period or reduced severity. Lower load
the bridge site. factors may be used to account for the acceptability of higher
temporary stress levels.
5.2—Dead loads
5.2.1 Structure dead loads—Structure dead loads are 5.4—Deformation effects
loads imposed on a member by its own weight and the 5.4.1 Settlement of supports—Stresses and forces
weight of other structural elements supported by it, including resulting from possible differential settlement of the
rails, sidewalks, slabs, and beams. If rails and sidewalks are supports should be considered in the design. Relaxation of
constructed after the main load-carrying system is self- such forces may be accounted for by using an effective
supporting, then the weight of these items is included in modulus of elasticity adjusted for time-dependent effects.5-3
superimposed dead loads.5-2 Preliminary design analyses may be made assuming
For the purpose of preliminary sizing of bridge members, differential settlements equal to a fraction of the average of
the unit structure dead load of typical concrete bridges may adjacent span lengths as follows:
be estimated from: Pile foundations 1/500
Spread footings
w = a + bL/100 On soil 1/1000
where On rock 1/2000
w = unit load in psf (kN/m2) Values used on the final design should be determined
a,b = constants from Table 5.2 from the project soils report or by consultation with the
L = span length in ft (m) geotechnical engineer.
5.2.2 Superimposed dead loads—Members should be 5.4.2 Shrinkage and creep—Shrinkage and creep are
designed to support the weight of superimposed dead loads concrete deformations, which occur over a length of time (up
including earth fill, wearing surface, stay-in-place forms, to 20 years). In general, creep and shrinkage are influenced by:
tracks and ballast, barrier walls, and railings, waterproofing, a. Mix design
signs, architectural ornamentation, pipes, conduits, cables, b. Water/cement ratio
and any other appurtenances installed on the structure. c. Aggregate properties
The following values may be used as a first estimate of d. Cement content
superimposed dead load: e. Maturity (age) of concrete
Acting on Highway Bridges f. Ambient relative humidity
Barrier railings (New Jersey curb) g. Reinforcing steel percentage
450 to 600 plf (6.6 to 8.8 kN/m) h. Size of member
Allowance for future overlay i. Shape of member
25 psf (1.2 KPa) j. Age at first application of load
Acting on Railroad Bridges k. Duration of load
Ballast and tracks
3500 plf (51.0 kN/m) Table 5.2—Constants for typical structure dead
Tracks only loads
600 plf (8.8 kN/m) Structure type Span range a b
Walkway 20-150 ft 53 530
Slab 6-16 m 2.54 0.8
500 plf (7.3 kN/m)
Tee-beam 35-80 ft 70 164
11-24 m 3.35 25.7
5.3—Construction, handling, and erection loads Slab and prestressed I-girders 50-130 ft 112 96
Consideration should be given to temporary loads caused 16-40 m 5.36 15.1
by the sequence of construction stages, forming, falsework, Prismatic box girders (falsework 80-160 ft 185 72
construction) 24-49 m 8.86 11.3
or construction equipment and the stresses created by lifting
Prismatic box girders (span-by-span 150-350 ft 285 40
and placing precast members. While one or more construc- construction) 46-107 m 13.65 6.3
tion schemes are generally considered when evaluating the Nonprismatic box girders (cantilever 350-800 ft 0 94
feasibility of the project, developing the actual construction construction) 107-244 m 0 14.8
343R-52 MANUAL OF CONCRETE PRACTICE
Fig. 5.4—Temperature distribution for design.5-7
Depth of the investigation of the effects of creep and
shrinkage depends upon the sensitivity of the structural
system to these effects. Structural systems for which creep
and shrinkage produce critical conditions should be investi-
gated for a range of deformations likely to occur. In choosing
the range of deformations, it should be kept in mind that the
designer does not have adequate control, or even adequate
knowledge, of all the conditions influencing creep and
shrinkage. The guidance which follows is of a general nature
and the designer is referred to ACI 209R and References 5-4,
5-5, and 5-6.
5.4.2.1 Shrinkage—Shrinkage is the reduction in
volume which occurs in concrete during the hydration
process due to loss of water not bound by hydration.
Shrinkage occurring before the concrete has taken its initial
set is not included in the shrinkage considered herein. The
practice of staggered placement of concrete can reduce
effects attributable to early plastic shrinkage, but is not effec-
tive in compensating for shrinkage which occurs over long
periods of time.
Increased volume changes resulting from the use of expan-
sive admixtures also should be considered in the design.
For preliminary design, ultimate shrinkage deformation in
conventional reinforced concrete bridges in moderate
climates, having a mean annual relative humidity between
50 and 90 percent, may be taken as a strain of 0.0002. The
shrinkage coefficient for lightweight concrete should be
determined by test for the type of aggregate to be used.
Rate of development of shrinkage may be approximated
by the following expression:
(ξsh)t = (ξsh)u t/(35 + t)
(ξsh)t = shrinkage of time t
(ξsh)u = ultimate shrinkage
t = time in days from end of curing (7 days for normal
curing, 1 to 3 days for steam curing)
This expression is based on a 6 in.- (150-mm) thick slab of
typical field-cast concrete, moist cured for 7 days, and
drying at 70 F (20 C). Corrections are applied for other
conditions. Correction for the thickness may be estimated5-6
by adjusting the time by the square of the rates of thickness
as follows
t* = t(h/ho)2
where
t* = equivalent time
t = actual time in days
h = actual thickness
ho = reference thickness, 6 in. (150 mm)
For more detailed information, the designer is referred to
ACI 209R.
5.4.2.2 Creep—Creep (time-dependent deformation under
constant load) and relaxation (time-dependent force under
constant strain) are influenced by the factors noted previ-
ously, but are most strongly influenced by the age of the
concrete (maturity) at the time of loading.
Creep deformation is approximately proportional to stress
for normal service loads. Creep may be characterized by the
creep coefficient, which is defined as the ratio of creep defor-
mation to instantaneous elastic strain. The ultimate creep
coefficient Cu for conventional reinforced concrete bridge
members,5-3 in a moderate climate and loaded at an age of 28
days, may be taken for preliminary design as
Cu = 3.0 – 0.030RH
where
RH = mean annual relative humidity
The development of creep deformation with time may be
approximated by the following expression
C(t) = Cu t0.6/(10 + t0.6)
where
t = age in days from loading
Creep strain at time t after loading is given by
(ξcr)t = (ξi)C(t)
where
(ξi) = instantaneous strain at application of load
The designer should remember that total strain includes
instantaneous strain, creep strain, and shrinkage strain. For
more detailed information, the designer is referred to ACI 209R.
5.4.3 Axial load deformations—The effect of differential
displacements at top of piers due to internal axial forces in
the main longitudinal members should be considered in the
analysis.
5.4.4 Thermal effects—Stresses or movements due to
temperature changes and to temperature differential in the
 BRIDGE ANALYSIS AND DESIGN
member should be considered in the design. The thermal
coefficient for normal weight concrete members may be
taken as 0.000006 F (0.000011 C). The temperature range
should be determined for the locality in which the structure
is to be constructed. In the absence of site-specific data, the
long-term change in temperature from the mean annual
temperature may be taken as given in AASHTO:
Temperature Temperature
rise fall
Moderate climate 30 F (17 C) 40 F (22 C)
Cold climate 35 F (19 C) 45 F (25 C)
Short-term (24- to 48-hr) temperature changes and solar
radiation produce differential temperatures through the depth
of a concrete bridge. Stresses produced by this nonuniform
temperature distribution can be significant at the service load
level. Such differential temperatures do not reduce the ulti-
mate load capacity. A temperature distribution appropriate for
calculation of the effects of differential temperature is shown
in Fig. 5.4. The temperature distribution for a given site and
structure may be derived by heat flow calculations on a section
of the structure for prescribed values of air temperature, solar
radiation, and thermal properties.
Stresses developed due to restraint of thermal expansion
may be conceptually divided into three parts:
1. Section restraint stresses—Stresses arising from
assumption that plane sections remain plane
2. Flexural restraint stresses—Stresses arising from the
support restraints of member end rotation (i.e., continuous
spans or fixed ends)
3. Axial restraint stresses—Stresses arising from support
restraint of the overall longitudinal expansion and contraction
These stresses can be calculated using the procedures
described in References 5-7 and 5-8.
5.4.5 Prestress effects—Stresses should be investigated
for the initial prestress force and for the prestress force
remaining after losses. Deformation of a member due to
prestressing should be allowed for in the design.
5.4.6 Frictional forces—The horizontal force due to fric-
tion of sliding plate bearings, roller bearings, pin, and rocker
bearings or stresses in elastomeric bearing pads should be
allowed for in the design. Typical design values for friction
are given in Table 5.4.6.
5.5—Environmental loads
5.5.1 Wind loads
5.5.1.1 Selection of procedures—The procedure
outlined in 5.5.1.2 may be used for conventional bridge
structures with span lengths not exceeding 400 ft (120 m).
These structures are generally sufficiently stiff so that
dynamic effects of wind need not be considered. Many
longer span bridges and some pedestrian, pipe, conveyor,
and sign bridges are sensitive to wind-induced oscillations.
For these structures, a more detailed analysis, including a
wind tunnel test, is recommended.5-2
In lieu of the procedure of 5.5.1.2, the simpler provisions
of 5.5.1.3 may be used for conventional girder and slab
bridges.
343R-53
Table 5.4.6—Coefficients of friction
Average coefficient of
Bearing surfaces static friction
Steel-steel 0.20
Steel-self lubricating bronze (Lubrite) 0.10
Stainless steel-TFE* (reinforced) 0.06
Stainless steel-TFE (virgin) 0.03
*
TFE = Polytetrafluoroethylene (Teflon).
5.5.1.2 General procedure—General procedures for
calculation of wind loads on bridges should consider clima-
tological conditions of the site, variation of wind speed with
height above the ground, gustiness of the wind, shape of the
structure, and its orientation to the mean wind direction and
natural frequencies of the structure. The following procedure
is based on Reference 5-1 and ANSI A58.1. The pressures
given are pressures against rigid structures. Effects due to
interaction between the dynamic characteristics of the struc-
ture and the wind are not included. These effects are not
usually significant for concrete bridges.
Basic wind velocity pressures may be determined from the
following formula
q = pV2/(2g) in psf (kPa)
where
q = dynamic wind pressure, psf (kPa)
V = maximum probable wind velocity, ft/sec (m/sec)
p = unit weight of air, 0.08 lb/ft3 at 32 F (12.6 N/m at 0 C)
g = acceleration due to gravity = 32.2 ft/sec2 (9.81
m/sec2)
The maximum probable wind velocity, V, should be
assumed to be not less than 100 mph (160 kph), unless a
lower value can be justified for the structure site. Permanent
features of terrain may allow a reduction of wind velocity at
various angles to the structure. Gust factors and the chan-
neling effect of roadways, valleys, and canyons should be
considered in determining the wind velocity at the structure
site. Gusts of 1.4 times the mean hourly wind velocities are
common for mean velocities in the range of 50 to 100 mph
(80 to 160 kph).
In the United States, Fig. 5.5.1 gives the basic wind speed
at a height of 33 ft (10 m) above the ground associated with
an annual probability of occurrence of 0.02 (50-year mean
recurrence interval).
Similar data are available in Reference 5-1 for the Cana-
dian provinces.
Wind load on the superstructure includes a horizontal load
Wh and a vertical load Wv, both acting directly on the struc-
ture, and WL acting on the live load.
The horizontal wind pressure Wh is applied to the total
exposed area of the structure as seen in elevation at right
angles to the longitudinal axis of the structure
Wh = Ce Cw q, in psf (kPa)
where
Ce = exposure coefficient
343R-54 MANUAL OF CONCRETE PRACTICE

Fig. 5.5.1—Annual extreme fastest-mile speed 30 ft above ground. 50-year mean recurrence interval.

Table 5.5.1a—Range of shape factors, Cw5-9 where z = height of the top of the superstructure above the
Longitudinal loads
ground or water in feet (meters). The exposure coefficient
Transverse loads 0 skew 60 skew
should not be taken as less than 1.0.
Girder superstructures 1.2 to 1.9 0.1 0.3 The vertical wind load consists of an upward or downward
Pony truss superstructures 1.6 to 1.8 0.1 0.5 pressure Wv over the total structure plan area applied at the
Through truss superstructures 2.4 to 2.7 0.1 1.7 windward quarter point of the transverse width of the structure

Cw = shape factor (Table 5.5.1a) Wv = Ceq

For the selection of shape factor values, consideration
should be given to the following factors:
Skew angle—Horizontal angle of wind to edge of structure.
Pitch angle—Vertical angle of wind to structure.
Aspect ratio—Ratio of length to width of structure.
Relative exposure—Members on the leeward side of
bridge may receive partial wind pressure.
Turbulence—Affected by size, shape, and positioning of
members.
For unusual structures, the shape factor, Cw, should be
determined by a wind tunnel test.
The exposure coefficient, Ce, accounts for the vertical
variation of mean wind speed, and is estimated by the
following formula
Ce = 0.50 z0.2 (0.65 z0.2)
The wind load on live load is treated as a uniformly
distributed load on the exposed area of the live load using the
bridge. The intensity of this load should be estimated using
the formula for Wh already given, with Cw = 1.2, and V not
exceeding 55 mph (90 kph). The exposed area should extend
over the entire length of the structure. Areas below the top of
a solid barrier or rail should be neglected.
For conventional concrete bridges, the wind load on live
load may be taken as a linear load of 100 plf (1.5 kN/m),
acting at 6 ft (1.8 m) above the top of the deck.
The wind loads acting on the superstructure are trans-
mitted to the substructure by the bearings or other appro-
priate devices. The horizontal wind loads should be divided
into transverse and longitudinal components by the skew
angle factors tabulated in Table 5.5.1b.
 BRIDGE ANALYSIS AND DESIGN
The horizontal wind load acting directly on the substruc-
ture should be calculated using the expression for Wh. The
shape factor should be 0.7 for circular cross sections, 1.4 for
octagonal, and 2.0 for rectangular shapes. This load is
applied in the direction of the wind. The possibility that a
skew wind direction is most critical should be considered.
5.5.1.3 Simplified procedure—In lieu of the general
procedure in 5.5.1.2, a simplified procedure may be used for
conventional concrete girder bridges not exceeding 200 ft
(60 m) in span. The loads specified are based on a wind
speed of 100 mph (160 kph). Where the probable maximum
wind speed can be reasonably ascertained, and is signifi-
cantly different from this base value, the specified loads may
be adjusted in proportion to the square of the wind speed.
a. For superstructure:
Wh = 50 psf (2.4 kPa) but not less than 300 lb per linear
foot (4.4 kN/m) of the structure span length
Wv = 20 psf (1.0 kPa)
WL = 100 plf (1.5 kN/m) acting at 6 ft (1.8 m) above the
deck
b. For substructure:
Forces from superstructure:
Wh = 50 psf (2.4 kPa) transverse and 12 psf (0.6 kPa)
longitudinal acting simultaneously
WL = 100 plf (1.5 kn/m) transverse and 40 plf (0.6 kN/m)
longitudinal acting simultaneously
Wv = 20 psf (1.0 kPa)
Forces on the substructure:
Wh = 40 psf (2.0 kPa)
5.5.2 Snowloads—Freshly fallen snow weighs from 5 to
12 lb/ft3 (0.8 to 1.9 kN/m3) while moistened snow compacted
by rain will weigh from 15 to 20 lb/ft3 (2.4 to 3.1 kN/m). Snow
partially thawed and then refrozen may approach the density
of ice. Because of this wide variation of weights, snow loads
should be determined from weather records and experience
at the site of the proposed structure.
Snow shelters provided over bridges should be designed
for any probable snow drift depth of the region. Snow loads
in excess of 20 psf (1.0 kPa) may be reduced by (SN-20)/40
for each degree that the shelter inclination with the hori-
zontal exceeds 20 deg (0.35 rad), where SN is the anticipated
snow load in psf (kPa).
Snow loads for unsheltered bridges are considered offset
by accompanying decreases in the live loads. The snow load
should be generally considered in the design in lieu of the
live load, whenever the anticipated snow load exceeds the
live load. Designers should also consider the weight of snow
removal equipment.
5.5.3 Earthquake loads—Earthquake ground motion
produces displacements of the structure relative to the
ground. These displacements are calculated from dynamic
analyses or equivalent static force analyses. For dynamic
analysis, the response of a given mode of structure vibration
is determined from the pseudo-velocity spectrum derived
from the earthquake ground motion. Response of one or
more normal modes of vibration are combined. For equiva-
lent static force analysis, the effects of the relative motion
between the ground and the structure are determined by
343R-55
Table 5.5.1b—Skew angle factor for wind loads on
substructures
Truss span Girder span
Transverse Transverse
horizontal or longitudinal Horizontal or Longitudinal
Skew angle vertical horizontal vertical horizontal
0 1.00 0.00 1.00 0.00
15 0.93 0.16 0.88 0.12
30 0.87 0.37 0.82 0.24
45 0.63 0.55 0.66 0.32
160 0.33 0.67 0.34 0.38
applying inertia loads to the structure. These inertia loads are
the product of structure weight and an effective acceleration
coefficient. Reduction of inertia forces due to yielding may
be either incorporated into the effective acceleration coeffi-
cient or considered separately. These parameters are influ-
enced by the seismicity of the site area, soil profile at the site,
and distance of the site from causative faults (if known). The
problem is complex, both from the standpoint of establish-
ment of the characteristics of the input motion and the analysis
of the response.5-10
Analysis of the response of the structure to a given input
motion may involve time-history analyses explicitly incor-
porating soil-structure interaction and yielding of various
members. Also, linear elastic modal response techniques are
commonly used. Yielding of various members is accounted
for by reducing the member forces determined from the
analysis by a factor dependent upon the assumed permissible
ductility. For simple structures in areas of moderate seismicity,
a pseudo-static analysis using a lateral load expressed as a
percentage of gravity may be adequate. Analysis of structure
response to seismic events is an analysis of imposed
deformations. It is characterized by forces only for conve-
nience of analysis.
Details should be designed to assure that the assumed
ductility can be achieved.5-10 Since plastic hinging will typi-
cally occur in columns for major seismic events, confinement
of column vertical reinforcing steel is a major concern.
Adequate length of bearing seats is of particular importance.
Bridges are generally long with respect to the wave length of
the predominant frequency of ground motions. Supports can
experience a movement that is out-of-phase with the
superstructure.
Structures may be classified according to their impor-
tance to the community and differing standards prescribed
accordingly.5-10
Simplification of Reference 5-10 is not possible without
significant loss of content. Designers should therefore consult
that reference for guidance in seismic design of bridges.
5.5.3.1 Historical—The first provision for earthquake
design criteria in codes for bridges in the United States was
established in the 1958 AASHTO Standard Specifications.
These specifications were based in part on the lateral force
requirements for buildings developed by the Structural Engi-
neers Association of California,5-11 as well as the equivalent
static force approach. The following provision applies to
bridges regardless of the location of the structures:
343R-56 MANUAL OF CONCRETE PRACTICE
V = kW
where
V = horizontal earthquake force
W = total weight
k = 0.02 for spread footings on soils having an allowable
bearing capacity greater than 3.5 tons/ft2 (171 kPa)
k = 0.04 for spread footings on soils having an allow-
able bearing capacity less than 3.5 tons/ft2 (171 kPa)
k = 0.06 for pile footings
Prior to the 1971 San Fernando earthquake, very little
damage to bridges was known to have resulted from seismi-
cally induced vibration effects. Most of the damage on a
worldwide basis was caused by permanent ground displace-
ment. The San Fernando earthquake represented a major
turning point in the development of seismic design criteria
for bridges in the United States.
In 1973, the California Department of Transportation
(CALTRANS) introduced new seismic design criteria for
bridges which included the relationship of the bridge site to
active faults; the seismic response of the soils at the bridge
site; and the dynamic response characteristics of the bridge.
In 1975, AASHTO adopted interim specifications which
were a slightly modified version of the 1973 CALTRANS
provisions and made them applicable to all regions of the
United States. In addition to these code changes, the 1972
San Fernando earthquake stimulated research activity on the
seismic problems related to the bridges. The Applied Tech-
nology Council (ATC), with funding from the Federal
Highway Administration (FHWA), developed recom-
mended seismic design guidelines for bridges. In 1983,
AASHTO adopted the ATC work as a guide specification.
5.5.3.2 ATC method—The ATC Seismic Specifications
were developed for four seismic performance categories
(SPC), A through D, to account for seismic risk varying from
very small, Category A, to very high, Category D. The SPC
is defined on the basis of both the acceleration coefficient for
the site and the importance of the bridge.
The ATC earthquake guidelines are based on a low prob-
ability of the design forces being exceeded during the normal
life expectancy of a bridge. Bridges and their components
that are designed to resist these forces and that are
constructed in accordance with the recommended design
details contained in the provisions may suffer damage, but
they will have a low probability of collapse due to seismi-
cally induced ground shaking.
The principles used for the development of the provisions
are:
a) Slight to moderate earthquakes should be resisted
within the elastic range of the structural components
without significant damage
b) Realistic seismic ground motion intensities and forces
are used in the design procedures
c) Exposure to shaking from severe earthquakes should
not cause collapse of the bridge. Where possible,
damage should be readily detectable and accessible for
inspection and repair.
Two different approaches were combined in the ATC
specifications to satisfy the preceding philosophy. They are
the “force design” and the “displacement control” criteria.
Minimum requirements are specified for bearing support
lengths of girders at abutments, columns, and hinge sections
to allow for important relative displacement effects that
cannot be calculated by current state-of-the-art methods.
Member design forces are calculated to account for the direc-
tional uncertainty of earthquake motions and for the simulta-
neous occurrence of earthquake forces in two perpendicular
horizontal directions. Design requirements and forces for
foundations are intended to minimize damage, since damage
to foundations would not be readily detectable.
The design displacement control criteria provided in ATC
requirements for minimum bearing support lengths at the
expansion ends of all girders are as follows:
Seismic Minimum support
performance length N
Category
A and B N = 8 + 0.02L + 0.08H (in.) Eq. (5-1)
or
N = 203 + 1.67L + 6.66H (mm) Eq. (5-2)
C and D N = 12 + 0.03L + 0.12H (in.) Eq. (5-3)
or
N + 305 + 2.5L + 10H (mm) Eq. (5-4)
N = minimum support length (in. or mm) measured
normal to the face of an abutment or pier
L = length, ft or m, of the bridge deck from the bearing
to the adjacent expansion joint, or to the end of the
bridge deck. For hinges within a span, L is the sum
of L1 and L2, the distances to either side of the
hinge. For single span bridges, L is the length of the
bridge deck. These lengths are shown in Fig.
5.5.3.2.
For abutments
H = average height, ft or m, of columns supporting the
bridge deck to the next expansion joint
H = 0 for single-span bridges
For columns and/or piers
H = column or pier height, ft, for Eq. (5-1) and (5-3)
or m, for Eq. (5-2) and (5-4)
For hinges within a span
H = average height of the adjacent two columns or
piers, ft, for Eq. (5-1) and (5-3) or m, for Eq. (5-
2) and (5-4)
5.5.3.3 AASHTO specifications—The 1983 AASHTO
Standard Specifications for Highway Bridges requires that
“in regions where earthquakes may be anticipated, structures
shall be designed to resist earthquake motion by considering
the relationship of the site to active faults; the seismic
response of the soils at the site; and the dynamic response
 BRIDGE ANALYSIS AND DESIGN 343R-57

characteristics of the total structure.” Restraining features

(hinge ties, shear blocks, etc.) should be provided to limit the
displacement of the superstructure. These restraints should
be designed to resist a force equal to 25 percent of the
contributing dead load less the column shears due to the
earthquake loading. Where an earthquake that could cause
major damage to construction has a high probability of
occurrence, lateral reinforcement shall be designed and
detailed to provide adequate strength and ductility to resist
expected seismic movements.
5.5.4 Earth pressures—Structural elements which retain
fills should be proportioned to withstand the earth pressures,
as determined by recognized soil mechanics methods, whether
such pressures result from static loadings or dynamic effects,
including those from seismic loadings. Under certain soil
conditions, dynamic effects should be considered for loads
during construction such as those resulting from compaction
procedures and heavy construction equipment.
When vehicular traffic can come within a horizontal
distance from the back of a structure or wall equal to one-half
its height, the earth pressure should be increased by a live
Fig. 5.5.3.2—Dimensions for minimum support length
load surcharge of not less than 2 ft (0.6 m) of earth.
requirements.5-10
Where an adequately designed reinforced concrete
approach slab supported at one end by the abutment is
provided, no live load surcharge need be considered on the 5.5.5 Buoyancy—Buoyancy should be considered, as it
back of the abutment. affects the design of the substructure, including piling, and
5.5.4.1 Active earth pressure—The active earth pressure of the superstructure. The adverse effects of buoyancy
used for the design of structures should be not less than an should be assumed to exist, unless water pressure can be
equivalent fluid pressure of 30 lb/ft3 (4.7 kN/m3). Values used positively excluded from beneath the foundation units.
in the United States for common well-drained backfill 5.5.6 Stream flow pressure—Stream flow pressure should
materials generally range from 35 to 40 pcf (5.5 to 6.3 kN/m3). be calculated by the formula
5.5.4.2 At rest earth pressure—Structures restrained
against movement or deflection under the action of earth SF = KV2 (0.515 KV2)
pressures should be designed to resist higher pressures than
given previously for active earth pressures. At rest pressures where
in granular materials may be estimated to be SF = stream flow pressure in psf (kPa)
V = velocity of water in ft/sec (m/s)
Po = γh/(1 + sin φ) K = constant, being 1.4 for square pier ends, 0.5 for
angled ends where the interior angle is 30 deg or
where less, and 2/3 for circular pier ends
γ = unit weight of soil 5.5.7 Ice loads—Piers and other portions of structures
h = depth of soil which are subject to the weight of ice, ice pressures, or the
φ = angle of internal friction force of floating ice should be proportioned accordingly. The
5.5.4.3 Passive earth pressure—Passive earth pressures designer should consult References 5-12 and 5-13 for guid-
should generally not be considered for the reduction of ance in design of structures subject to ice loading.
effects of other design loads. 5.5.7.1 Ice buildup on structures—Cables, hangers,
Expansion due to temperature rise of structures that do not guys, and their supporting members should be designed to
contain deck expansion joints may be sufficient to develop withstand the weight of an ice coating having a thickness
passive earth pressures on the end walls. End walls of such dependent upon the severity of the climate. This coating
structures should be designed to resist the passive earth should not be less than 0.5 in. (13 mm) nor greater than 1.5
pressures of the backfill material. in. (40 mm), which might occur in northern Texas, eastern
5.5.4.4 Drag force—Friction piles driven through fills, Oklahoma, Kansas, Illinois, Indiana, Iowa and states north of
or other materials that settle during the life of the structure, Pennsylvania. Wind pressures should be assumed to act on
should be designed to resist a drag force caused by negative these increased surface areas.
friction unless the force is positively eliminated by isolating Ice buildup on decks, beams, and truss members may be
the piles from the settling soils. In lieu of a more accurate considered to be offset by an accompanying decrease in
analysis, the drag force may be considered equal to the live loads and need not be included, except under special
weight of unconsolidated earth enclosed by the pile group. conditions.
343R-58 MANUAL OF CONCRETE PRACTICE
Table 5.5.7.2—Indentation coefficient
Aspect ratio b/h Ca
0.5 1.8
1.0 1.3
1.5 1.1
2.0 1.0
3.0 0.9
4.0 or greater 0.8
5.5.7.2 Dynamic ice pressures—Horizontal ice forces
resulting from the pressure of moving ice should be calcu-
lated by the formula
F = 144 CaCi σhb
where
F = horizontal ice force on pier, lb (kN)
Ca = indentation coefficient, see Table 5.5.7.2
Ci = coefficient for pier inclination from vertical cos(3α)
α = angle of pier inclination from vertical, deg
σ = effective ice strength as specified in the following,
psi (kPa)
h = thickness of ice in contact with pier, ft (m)
b = width or diameter of pier at level of ice action, ft (m)
Effective ice strength is normally taken in the range of 60
to 200 psi (400 to 1500 kPa) on the assumption that the ice
crushes on contact with the pier. The values used should be
based on an evaluation of the probable conditions of the ice
at time of movement, on previous local experience, and on
an assessment of an existing structure performance. The
following guide to values of effective ice strength appro-
priate to various situations may be used:
1) σ = 60 psi (400 kPa) where breakup occurs at melting
temperatures and where the structure of the ice has substan-
tially disintegrated and the ice flows as small “cakes.”
2) σ= 100 psi (700 kPa) where breakup occurs at melting
temperatures, but the ice is internally sound and moves in
large pieces.
3) σ = 150 psi (1100 kPa) where at breakup there is an
initial movement of the ice sheet as whole or where large
sheets of sound ice may strike the piers.
4) σ = 200 psi (1500 kPa) where breakup of major ice
movement may occur with ice temperature significantly
below the melting point.
Due consideration should be given to the probability of
extreme rather than average ice conditions at the site in question.
Piers should be placed with their longitudinal axis parallel
to the principal direction of ice action. The horizontal ice
force F calculated by the formula should be taken to act
along the direction of the longitudinal axis with a transverse
force not less than 15 percent of the horizontal force, acting
simultaneously.
Where the longitudinal axis of the pier cannot be placed
parallel to the principal direction of ice action, or where the
direction of ice action may shift, the total force on the pier
should be calculated by the formula, taking the width of the
structure to be that normal to the longitudinal axis of the pier.
The resulting force, taking into account its assumed direc-
tion of action, should then be resolved into vector compo-
nents parallel to and transverse to the longitudinal axis of the
pier. In such conditions the transverse force should not be
less than 20 percent of the total force.
In the case of slender and flexible piers, consideration
should be given to the oscillatory ice forces associated with
dynamic ice action and to the possibility of high momentary
forces, structural resonance, and fatigue.
5.5.7.3 Static ice pressure—Ice pressure on piers frozen
into ice sheets on large bodies of water should receive
special consideration where there is reason to believe that the
ice sheets are subject to significant thermal movements rela-
tive to the piers.
Design loads to pier sides and ends due to ice pressures
from static ice sheets should not be less than 10,000 lb/linear
ft (146 kN/m). This ice pressure should be assumed contin-
uous or discontinuous on either side of pier as required to
produce maximum stresses, but need not be applied simulta-
neously with loads to pier noses from floating ice.
5.5.8 Debris loads—Loads from impact of floating debris
against piers or the superstructure should be considered in
the design. Debris buildup against piers can also signifi-
cantly increase streamflow forces. These loads should be
established on the basis of local knowledge of the character
of the stream. Debris loads should be included as environ-
mental loads for combination with other loads.5-12
5.5.9 Wave action—Loads against piers and superstructure
induced by wave action may be significant in exposed lake or
sea coast locations. Reference 5-14 should be consulted for
determination of wave heights and forces on structures.
5.5.10 Ship impact—The possibility of ship impact should
be considered in the design of bridges crossing navigable
waters.5-15 If the risk of impact warrants protective
measures, then piers should be protected from impact or
designed to resist the maximum likely impact.
Factors to be considered in assessing the risk of impact and
the maximum likely impact include:
Vessel types and traffic volumes
Bridge clearance, horizontal and vertical
Navigational aids, including radar reflectors
Interactions between vessels, current, and wind
Protective devices include:
Dolphins and fenders
Protective embankments
Signal lights may be provided on the bridge to stop motor
vehicle traffic in the event of a collision endangering the span.
5.6—Pedestrian bridge live loads
5.6.1 Deck live load—Pedestrian bridges should be
designed for a live load of 100 lb/ft2 (5.0 kN/m2) of walkway
area between curbs or rails. This uniform load may be
applied continuously or discontinuously over both length
and width of structure to produce the maximum stresses in
the members under consideration. For unusually wide and
long bridges, special load reduction may be used.
The maximum live load deflection of the span should prefer-
ably not exceed one-thousandth of the span length. Structures
 BRIDGE ANALYSIS AND DESIGN
of the suspension type and especially the “ribbon” bridge-
type5-16 should have a natural frequency that is higher than that
of the pedestrian traffic, to avoid any possible effect of reso-
nance. A natural frequency of vibration, in hz, greater than
f= 600 ⁄ l ( 183 ⁄ l )
where l is the span length, ft (m), is desirable.5-17
5.6.2 Railing live load—The minimum design loading for
pedestrian railing should be transverse and vertical uniform
loads of 50 lb/linear ft (0.75 kN/m), acting simultaneously
on each longitudinal member. Rail members located more
than 5 ft (1.5 m) above the walkway are excluded from these
requirements.
Posts should be designed for the transverse load acting at
the center of gravity of the upper rail, or for high rails, at 5 ft
(1.5 m) minimum above the walkway.
5.6.3 Provision for overload—Pedestrian bridges wide
enough for vehicles should be designed for an infrequent
load equal to the heaviest vehicle or equipment that is likely
to be moved over the bridge, but not less than two 4000-lb
(17.8-kN) axles spaced at 12 ft (3.6 m). The overload
assumed should be listed on the plans.
5.7—Highway bridge live loads
5.7.1 Standard vehicular live loads—Design vehicular
live loads should be developed from the vehicles actually in
use or normally allowed on the highways, taking into consid-
eration the possibilities of future higher legal loads and
permissible occasional overloads.
In the United States, the standard design vehicular live load,
together with the associated impact, centrifugal, and braking
forces, curb, and railing loads, special alternate live load, and
provision for overload, are contained in the “Standard Speci-
fications for Highway Bridges” published by the American
Association of State Highway and Transportation Officials.
The province of Ontario, based on a study of the actual and
legally permitted loadings on the highways, has developed a
vehicular configuration considerably more severe than the
AASHTO standard. The Ontario standard vehicular loads,
with associated curb and railing loads, centrifugal and
braking forces, dynamic effects, and performance factors,
are contained in the Ontario Highway Bridge Design Code.5-1
Background information on the development of the Ontario
standard loading can be found in References 5-18 and 5-19.
The loads specified in AASHTO and the Ontario code are
applicable for the design of ordinary highway bridges with
span lengths up to 400 ft (120 m).
5.7.2 Special truck loads
5.7.2.1 Logging trucks—The logging industry, particu-
larly in the western United States, makes use of specially
constructed roads for trucking equipment weighing consid-
erably greater than allowed on the state highway systems.
Bridges for logging roads should be designed for the
trucking equipment and actual loads expected.
5.7.2.2 Military loads—For design of bridges for mili-
tary loads or review of existing bridges for passage of mili-
343R-59
tary loads, information should be obtained from the
appropriate military manual.5-20
5.7.2.3 Overload provisions—Provision should be made
in the design for infrequent heavy vehicle loads. If the nominal
design live load for which the structure is designed is less than
the standard H20 loading, the design vehicle increased by 100
percent should be applied in any single lane without concur-
rent loading of any other lanes. The overload should apply to
all parts of the structure affected except deck slabs.
Structures on major arterial streets or highways should be
designed for specific overload of “permit” vehicles oper-
ating either without restrictions as to location, speed, and
concurrent traffic or under police supervision. A typical
vehicle for which unrestricted permits are issued is illus-
trated in Fig. 5.7.2.3. Combination of overload vehicle loads
with other loads is discussed in Section 5.12.
5.7.3 Application of vehicular live loading
5.7.3.1 Design traffic lanes—For the purpose of design,
the bridge roadway is divided into longitudinal design traffic
lanes of equal width. The width of each lane depends upon
the total width of roadway between curbs, and should be
determined in accordance with the provisions of the relevant
design specifications.
5.7.3.2 Traffic lane units—The standard vehicular live
loading, whether truck loading or lane loading, occupies a
width of 10 ft (3 m). In computing stresses, each lane loading
or individual truck should be considered as a unit. Fractional
load lane widths should not be used.
5.7.3.3 Positioning of live loads—The numbers and
position of the vehicular loadings should be such as to
produce the maximum effect on the member. Specific rules
concerning the application of live loading, particularly with
regard to continuous spans, are given in relevant design
specifications.
5.7.4 Reduction in load intensity—Where maximum
stresses are produced in any member by loading several
traffic lanes simultaneously, the following percentages of the
live loads should be used in view of the improbability of
coincident maximum loading:
Percent
First and second lanes 100
Third lane 70
Fourth lane, and subsequent lanes 40
The reduction in intensity of loads on transverse members
such as floor beams should be determined, as in the case of
main trusses or girders, using the number of traffic lanes
across the width of roadway which are loaded to produce
maximum stresses in the floor beams.
5.7.5 Distribution of loads to beams—Refer to Chapter 10.
5.7.6 Fatigue—In the design of concrete highway bridges,
fatigue should be considered in design of reinforcing steel
and prestressing steel in cracked sections. Steel requirements
in deck slabs may be governed by fatigue. The extreme
wheel or axle load will be the governing load for slab design.
The number of load repetitions can be estimated from the
estimated average annual daily truck traffic for the bridge. If
the main longitudinal steel is to be investigated for fatigue,
343R-60 MANUAL OF CONCRETE PRACTICE
Fig. 5.7.2.3—Permit design live loads.
then it may be appropriate to consider an equivalent design
vehicle for fatigue, as set forth in the next paragraph.5-2
The expected distribution of truck traffic should be repre-
sented by a fatigue design truck that is the same as the HS20-
44 truck, except that its gross weight W should be selected,
so that the number of cycles to failure for the fatigue design
truck is the same as the total number of cycles to failure for
the different trucks in the distribution. If specific information
is available on the expected distribution of truck traffic, Wf
should be determined from
3 3
Wf = Σi ai wi Σi ai = 1
in which ai = the fraction of trucks with a weight Wi. If
specific information is not available on the expected distri-
bution of truck traffic, Wi should be equal to 50 kips (220
kN). If specific information is available on truck traffic
volume, each lane of bridges on two-lane highways should
be designed for the total truck traffic volume in both direc-
tions and each lane of bridges on highways of more than two
lanes should be designed for the truck traffic volume in one
direction. If specific information on the truck traffic volume
is not available, each lane of bridges on major rural highways
should be designed for an annual average daily truck traffic,
AADTT, of not less than 400 times the total number of lanes,
and each lane of bridges on major urban highways should be
designed for an AADTT of not less than 600 times the total
number of lanes. As used previously, AADTT excludes panel,
pickup, and two-axle/four-wheel trucks, but includes buses.
5.8—Railroad bridge live loads
5.8.1 Design live loads—Design railway loads have been
developed from engines and gross car loads which have been
used. Cooper E80 loading is recommended for each track of
main line structures. This train loading, together with the
associated impact, centrifugal, lateral, and longitudinal
forces, is fully detailed in Chapter 8 of the Manual of the
American Railway Engineering Association.
5.8.2 Provisions for overload—Allowance should be
made for occasional overloads on the structure. The antici-
pated overload, including impact, when substituted for the
design live load and impact, should not result in stresses that
exceed the allowable stresses by more than 50 percent.
Where provisions are made for overloads to cross the
structure at reduced speeds, the impact load for speeds less
than 40 mph (60 kph) may be reduced in a straight line vari-
ation from full impact at 40 mph (60 kph) to 0.2 of the full
effect at 10 mph (15 kph).
5.9—Rail transit bridge live loads
A comprehensive report on rail transit bridge live loads
has been issued by ACI Committee 358. Loadings are clas-
sified according to heavy rail transit (similar) to railway
loadings and light rail transit. Loadings are specific to each
project, depending upon the equipment chosen.
5.10—Airport runway bridge loads
5.10.1 Landing gear loads—Design landing gear loads for
runway and taxiway bridges vary according to the type of
service anticipated for the airport. Because of the increased
use of larger, heavier aircraft for all types of service, struc-
tures should be designed for the heaviest aircraft probable at
the airport. Table 5.10 lists the landing gear loads and dimen-
sions of several airplanes presently in use throughout the
world. A more complete list of aircraft dimensions and
weights is provided in the Federal Aviation Agency’s Advisory
Circular, Aircraft Data.5-21 Definitive data for dimensions
and weights of aircraft are available from manufacturers in
standardized publications entitled Airplane Characteristic,
Airport Planning. Fig. 5.10 shows general landing gear
patterns for a variety of current commercial aircraft. Wheel
spacing and spacing of axles in tandem groups is given in
Table 5.10. For definite dimensions, see manufacturer's data
on specific aircraft. Continental and intercontinental class
airports should be designed to accommodate aircraft similar
in size and weight to the Boeing 747.
5.10.2 Impact—Live loads produced by aircraft landing
gear should be increased by the following percentages to
account for vibration and impact effects on portions of the
structure listed in Group A following.
Parking aprons and low-speed taxiways 30 percent
High-speed taxiways and runways 40 percent
Touch-down areas of runways 100 percent
The tire contact areas may be increased accordingly.
Impact need not be considered for portions of the structure
listed in Group B. Impact for structures covered with fill
varies proportionally from the percentages shown for no fill
to 0 percent for 10 ft (3 m) of fill.
Group A:
1) Superstructure, columns, and pedestals which support
the superstructure with rigid, fixed, or expansion bearings, or
which are rigidly attached to the superstructure, and legs or
rigid frames
 BRIDGE ANALYSIS AND DESIGN 343R-61

Table 5.10—Aircraft loads and wheel spacings

Main landing gear Each gear
(1) (2) (3) (4) (5) (6) (7)

Aircraft x, tandem spacing, Contact area per Maximum gear Maximum gross
model/series Type of gear y, dual spacing, in. in. tire, in.2 load, lb Number of wheels weight, lb
A300 B, B2 Dual-tandem 35 55 207 138,800 4 304,000
A300 B, B4 Dual-tandem 36.2 55 197 151,900 4 333,000
B52 Dual-dual 37 99 295 231,000 4 462,000
B707-320 B, C Dual-tandem 34.6 56 218 152,000 4 320,000
B720 Dual-tandem 32 49 188 109,950 4 230,000
B727-100 Dual 34 — 233 73,500 2 154,700
B727-100C Dual 34 — 233 76,900 2 161,900
B727-200 Dual 34 — 233 79,900 2 168,200
B727-Adv. 200 Dual 34 — 295 86,800 to 96,800 2 183,000 to 204,000
B737-100 Dual 30.5 — 160 51,000 2 107,000
B737-200 Dual 30.5 — 160 54,000 2 113,700
B737-Adv. 200 Dual 30.5 — 160 59,000 2 124,000
B737-300 Dual 30.5 — 160 62,000 2 130,500
B747 Dual-tandem* 44 58 204 166,500 4 778,000
B747 SP Dual-tandem* 43.25 54 196 147,400 4 653,000
B757-200 Basic Dual-tandem 34 45 159 102,900 4 221,000
B757-200 Option 1 Dual-tandem 34 45 159 105,600 4 231,000
B757-200 Option 2 Dual-tandem 34 45 159 108,900 4 241,000
B767 (1) Dual-tandem 45 56 178 132,000 4 282,000
B767 (2) Dual-tandem 45 56 210 142,600 4 302,000
BAC 1-11 Dual 20.38 — 145 50,800 2 107,000
C5A Twin-triangular† (4 groups) 230 165,000 6 763,000
C130 Single-tandem — 60 356 71,300 2 150,000
C131 Dual 21.7 — 130 28,700 2 60,500
C133B Dual-tandem 30 75.6 341 142,400 4 300,000
C135 Dual-tandem 36 60 230 132,000 4 278,000
C141 Dual-tandem 32.7 50.4 203 151,000 4 318,000
Concorde Dual-tandem 26.4 66 250 184,300 4 388,000
DC3 Single — — 266 11,970 1 25,200
DC7C Dual 30 — 272 67,900 2 143,000
DC8-10,20 Dual-tandem 30 55 224 130,900 4 275,600
DC8-30,40,50 Dual-tandem 30 55 213 149,700 4 315,300
DC8-55,61 Dual-tandem 30 55 209 155,800 4 328,000
DC8-62 Dual-tandem 32 55 220 167,700 4 353,000
DC8-63 Dual-tandem 32 55 220 170,100 4 358,000
DC9-10 Dual 24 — 161 43,080 2 90,700
DC9-20 Dual 24 — 161 47,500 2 100,000
DC9-30 Dual 24 — 161 51,300 2 108,000
DC9-40 Dual 24 — 161 54,150 2 114,000
DC9-50 Dual 24 — 161 57,480 2 121,000
DC9-80-81 Dual 28.12 — 195 67,500 2 141,000
DC9-80-82 Dual 28.12 — 195 71,500 2 150,500
DC9-80-83 Dual 28.12 — 195 75,000 2 161,000
DC10-10 Dual-tandem 54 64 254 189,300 4 413,000
DC10-20,30 main Dual-tandem 32 55 254 211,300 4 510,000
gear
DC10-20,30 center Dual 38 — 276 87,500 2 510,000
gear only
KC135A Dual-tandem 36 60 250 145,000 4 312,000
L500 Dual-tandem 53 72 240 200,000 4 421,000
L1011 Dual-tandem 52 70 278 195,000 4 410,000
*
B747 main undercarriage has four groups of four wheels each.
†
C5A main undercarriage has four groups of six wheels each, wheel coordinates for each gear are:

Wheel x y
1 0.00 0.00
2 0.00 –36.40
3 0.00 111.20
4 0.00 147.60
5 64.00 32.47
6 64.00 78.72
343R-62 MANUAL OF CONCRETE PRACTICE
Fig. 5.10—Aircraft landing gear arrangements (terms “twin” and “dual” are used interchangeably).
2) The portion above the ground line of piles which are
rigidly connected to the superstructure as in rigid frame
structures
Group B:
1) Abutments, retaining walls, piers, pile caps, and piling
which are not rigidly connected to the superstructure
2) Buried foundation units, footings, and supporting soil,
and structures with 3 ft (1 m) or more of earth cover
5.10.3 Application of wheel loads
5.10.3.1 Slabs—Design wheel loads for slabs are equal to
the tire pressures uniformly distributed over the tire contact
area for each tire of the design aircraft. The width of transverse
slab resisting the load effects of an individual (or dual) tire
should not be assumed greater than the slab span, the spacing
of tandem axles, or 7 ft (2.1 m), whichever is less.
The width of longitudinal slab resisting the load effects of
an individual tire should not be assumed greater than the slab
span, the tire spacing, or 7 ft (2.1 m), whichever is less. The
resisting width for dual, dual-tandem or dual-dual tandem
landing gear may be twice the transverse spacing of the ties
in one gear, but should not exceed the slab span or the
spacing of landing gear units.
5.10.3.2 Beams and girders—Due to the proximity of
wheels, the total load of a landing gear unit may be assumed
as a single rolling load when considering the effect on longi-
tudinal or transverse beams and girders. The width of this
single rolling load may be assumed equal to the distance out
to out of tandem tires. Distribution of this load to longitu-
dinal or transverse beams and girders may be based on the
slab acting as a simply supported span between beams or
girders or the distribution factors for highway bridge loads
may be used.
When designing for dual-dual tandem landing gear such as
used on the Boeing 747, all girders, if monolithic with slab
between outside landing gear units, may be considered
equally effective in resisting the total aircraft load.
The load distribution procedures listed previously are
satisfactory for preliminary analysis or for the design of
minor structures. It is, however, recommended that the
design of major taxiway and runway structures be based on
the theoretical elastic analysis of load distribution or model
tests of the proposed structures.
5.10.4 Braking forces—Substructure units and the
connections between superstructure and substructure units
should be designed to resist a longitudinal braking force
equal to the following percentage of gross load of one
aircraft, without impact, unless more exact braking informa-
tion is available for the design aircraft. The designer should
check the manufacturer's data for specific aircraft.
Parking aprons and low-speed taxiway 30 percent
High-speed taxiways and runways 80 percent
5.10.5 Provisions for crash landing loads—Structures
located on portions of the runway where crash landings may
be anticipated should be designed for the failure load of the
landing gear.5-21 The Federal Aviation Agency requires that
landing gears have a reserve strength capable of withstanding
the impact of the aircraft dropping at a rate of 12 ft/sec
(3.6 m/sec).
5.11—Pipeline and conveyor bridge loads
5.11.1 Fluid loads—Pipeline bridges should be designed
to carry the heaviest fluid anticipated in the pipeline. Where
a concrete pipe (or channel) acts as a structural member of
the bridge, or is not specifically isolated from becoming
stressed longitudinally by the live load deflection of the
structure, stresses should be checked and reinforcement or
prestressing added as necessary.
If pipeline bends occur on a structure, the load from the
thrust blocks should be distributed to the substructure units,
taking into account the stiffness of the units and the rigidity
of the connection between the load and the units.
5.11.2 Solid loads—Structural members of ore bridges
and other structures designed to carry solid or semifluid
loads should be proportioned for a continuous or discontin-
uous loading, whichever produces the higher stresses. Forces
 BRIDGE ANALYSIS AND DESIGN
imposed by starting and stopping the conveyor should be
considered acting with the live load.
5.11.3 Equipment loads—The effect of maintenance
equipment on pipeline and ore bridges should be included
with or in lieu of the live loads.
Catwalks should be designed for a live load of not less
than 60 psf (3kN/m2) and the load should be either contin-
uous or discontinuous over the walkway, whichever
produces the higher stresses.
5.12—Load combinations
The loading groups listed in Table 5.12a and Table 5.12b
represent the basic combinations of loads and forces to
which a concrete bridge may be subjected. Each part of a
bridge, and the foundation on which it rests, should be
proportioned to resist the group loads applicable for the type
of structure and the particular site. If, in the engineer's judg-
ment, predictability of load conditions or materials of
construction are different than anticipated by the specifica-
tions, these should be accounted for by appropriate changes
in the allowable stresses, or load factors.
Service load design should be based on the percentage of
basic allowable stress indicated, except no increase in allow-
able stresses should be permitted for members or connec-
tions carrying wind loads only. When designing rigid frame
bridges, the condition that the lateral earth pressure could be
less than predicted should be investigated using a load factor
of 0.5 in loading group I. For this condition the percentage of
allowable stress may be taken at 125 percent.
Variable loads, such as live, buoyancy, and wind, should
be reduced or eliminated from the group whenever such
reduction or elimination will result in a more critical loading
for the member under consideration. When loading condi-
tions under consideration produce stresses of opposite sign
to the dead load stresses, or when the stability of the structure
may be impaired with reduced dead load, the calculated dead
load may be reduced as much as 25 percent.
Table 5.12a and Table 5.12b
Load factors
Notes
1. Permanent loads are loads which will be present on the
structure due to the nature of the design and construction.
Time-varying loads such as creep, shrinkage, and forces
induced due to differential support displacements are
assumed to be taken at a value appropriate to the time under
consideration (i.e., adjusted for relaxation).
2. Where it produces a more critical condition, the lower
value of earth pressure should be used. Factors for strength
design methods are intended to account for future changes in
the character of the backfill material.
3. Creep, shrinkage, and differential support displace-
ments are time-varying restraint forces which should be
adjusted for relaxation. They should be calculated using
stiffness properties that take cracking into account. The
double value factors for ultimate load represent the uncer-
tainties in the calculated values. The factor value creating the
most critical condition should be used.
343R-63
4. Buoyancy and streamflow should be based on the water
level appropriate to the other conditions being considered as
acting simultaneously.
5. Nosing and lurching are for railroad bridges only.
6. Construction load is considered as part of a separate
group (Group IV). If assumptions of construction loads are
made in the design, they should be noted on the plans. If the
contractor is required to provide stress analysis calculations
to prove the suitability of the construction method and equip-
ment, these requirements should be spelled out in detail in
the plans.
7. In general only one environmental load is to be applied
at a time except that wind on live load is considered in asso-
ciation with a reduced wind on the structure and ice may be
considered to act concurrently with high water if appropriate
to the site.
8. Wind for Group IV combination is assumed to act on
any equipment left on the structure for extended periods.
9. These factors assume that the earthquake forces used are
those associated with an event having a 90 percent proba-
bility of not being exceeded within 50 years.
Equations to be satisfied in design
Service load method
n
∑ αi f i ≤ βf a
i=1
where
n = number of individual loads in the load combina-
tion considered
αi = load factor for ith loading from Table 5.12a
fi = stress produced by ith loading
β = percent basic allowable stress from Table 5.12a
fa = basic allowable stress from Chapter 8
Strength method
n
∑ αi Pi ≤ φ Rn
i=1
where
n = number of individual loads in the load combination
considered
αi = load factor for the ith loading from Table 5.12
Pi = moment, shear, or axial load from the ith loading
φ = capacity-reduction factor from Section 7.2 or
Section 9.2.1
Rn = characteristic strength (moment, shear, axial load)
RECOMMENDED REFERENCES
The documents of the various standards-producing organi-
zations referred to in this report are listed here with their
serial designation, including year of adoption or revision.
The documents listed were the latest effort at the time this
report was written. Since some of these documents are
revised frequently, generally in minor detail only, the user of
this report should check directly with the sponsoring group
if it is desired to refer to the latest revision.
343R-64 MANUAL OF CONCRETE PRACTICE

Table 5.12a—Load factors

Design method Service load
Load group, percent basic stress I 100 perm. + trans. II 125 perm. + env. III 140 + env. IV 133 const.
Load cases Notes
Permanent loads
Structural DL D 1 1.0 1.0 1.0 1.0
Superimposed DL D 1.0 1.0 1.0 1.0
Earth pressure E 2 1.0/0.5 1.0/0.5 1.0/0.5 1.0/0.5
Creep S 3 1.0 1.0 1.0 1.0
Shrinkage S 3 1.0 1.0 1.0 1.0
Displ. of supts. DS 3 1.0 1.0 1.0 1.0
Buoyancy B 4 1.0 1.0 1.0 1.0
Transient loads
Live load L 1.0 1.0
Impact I 1.0 1.0
Longit. force LF 1.0 1.0
Centrif. force CF 1.0 1.0
Nosing N 5 1.0 1.0
Construction C 6 1.0
Environmental loads 7
Wind W 8 1.0 0.3 0.5
Wing on LL WL 1.0
Temperature T 1.0 1.0
Earthquake EQ 9 0.5
High water SF 1.0 1.0
Ice B 1.0

Table 5.12b—Load factors

Design method Ultimate strength
Load group I II III IV
Load cases Notes
Permanent loads
Structural DL D 1 1.3 1.3 1.3 1.3
Superimposed DL D 1.3 1.3 1.3 1.3
Earth pressure E 2 1.3/0.35 1.3/0.35 1.3/0.35 1.3/0
Creep S 3 1.3/0.7 1.3/0.7 1.3/0.7 1.3/0
Shrinkage S 3 1.3/0.7 1.3/0.7 1.3/0.7 1.3/0
Displ. of supt. DS 3 1.3/0.7 1.3/0.7 1.3/0.7 1.3/0
Buoyancy B 4 1.0 1.0 1.0 1.0
Transient loads
Live load L 2.2
Impact I 2.2
Longit. force LF 2.2
Centrif. force CF 2.2
Nosing N 5 2.2
Construction C 6 1.7
Environmental loads 7
Wind W 8 1.3 0.4 0.7
Wing on LL WL 1.3
Temperature T 1.3 1.3
Earthquake EQ 9 1.0
High Water SF 1.3 1.3
Ice B 1.3

Title F = Frictional forces

A = Axial load deformations and rib shortening I = Impact due to live load
B = Buoyancy ICE = Ice pressure
C = Construction, handling, and erection loads (add to Group L = Live load
II) LF = Longitudinal force from live load
N = Nosing and lurching forces
CF = Centrifugal force
OL = Overload
D = Dead load, including effects of prestressing force, if any S = Shrinkage and other volume changes
DR = Derailment force (to be included with “overload” in load- SF = Stream flow pressure
ing groups SN = Snow load (substitute for W in Group II
DS = Displacement of supports T = Temperature
E = Earth pressure W = Wind load
EQ = Earthquake (substitute for W in Group II) WL = Wind load on live load
 BRIDGE ANALYSIS AND DESIGN
American Association of State Highway and Transportation
Officials
GSDB Guide Specification for Seismic Design of
Highway Bridges, 1983
HB-13 Standard Specifications for Highway Bridges,
Thirteenth Edition, 1983
LTS-1 Standard Specifications for Structural Supports
for Highway Signs, Luminaires, and Traffic
Signals, 1975.
American Association of Railway Engineers
Manual for Railway Engineering, Chapter 8
American Concrete Institute
209R-82 Prediction of Creep, Shrinkage, and Tempera-
ture Effects in Concrete Structures
358R-80 Report on Concrete Guideways
American National Standards Institute
58.1-1982 Building Code Requirements for Minimum
Design Loads in Buildings and Other Structures
CITED REFERENCES
5-1. Ontario Ministry of Transportation and Communica-
tions, “Ontario Highway Bridge Design Code and Commen-
tary,” Downsview, Ontario, 1983.
5-2. “Recommended Design Loads for Bridges,” Journal
of the Structural Division, ASCE, V. 107, No. St. 7, July
1981, pp. 1161-1213.
5-3. Branson, D. E., Deformation of Concrete Structures,
McGraw-Hill Book Company, New York, 1977.
5-4. Comite Euro-International du Beton (CEB), “CEB-
FIP Model Code for Concrete Structures,” International
System of Unified Standard Codes of Practice for Structures,
V. II, 1978.
5-5. Meyers, B. B., and Branson, D. E., “Design Aid for
Predicting Creep and Shrinkage Properties of Concrete,”
ACI JOURNAL, Proceedings, V. 69, No. 9, Sept. 1972, pp.
551-555.
5-6. Bažant, Z. P., and Wittmann, F. H. (Ed.), Creep and
Shrinkage in Concrete Structures, John Wiley and Sons,
Inc., New York, 1982.
5-7. Priestley, M. J. N., “Design of Concrete Bridges for
Temperature Gradients,” ACI JOURNAL, Proceedings V. 75,
No. 5, May 1978, pp. 209-217.
343R-65
5-8. Elbadry, M. M., and Ghali, A., “Nonlinear Tempera-
ture Distribution and its Effect on Bridges,” IABSE
Proceedings, Zurich, 1983, pp. 66-83.
5-9. Highway Research Board Report 10, “Investigation
of Wind Forces on Highway Bridges,” 1953, National
Academy of Sciences, National Research Council, Publica-
tion 272.
5-10. “Seismic Design Guidelines for Highway Bridges
(ATC-6),” Applied Technology Council, Berkeley, Cali-
fornia, 1981.
5-11. Seismology Committee, Structural Engineers Asso-
ciation of California, “Recommended Lateral Force
Requirements,” July 1959, revised Dec. 1959.
5-12. Bennett, Iven, “Glaze, Its Meteorology and Clima-
tology, Geographical Distribution and Economic Effects,”
Headquarters Quartermaster Research and Engineering
Command, U.S. Army Quartermaster Research and Engi-
neering Center, Natick, Massachusetts, Environmental
Protection Research Division, Technical Report EP-105,
Mar. 1959.
5-13. “Ice Loads on Bridge Piers,” Structural Engineering
Series, No. 1, Jan. 1976, prepared by the Hydraulics and
Review Branches, Federal Highway Administration, Wash-
ington, D.C.
5-14. Shore Protection Manual, V. I, U.S. Army Coastal
Engineering Research Center, Ft. Belvoir, Va., 1973.
5-15. “Ship Collision with Bridges and Offshore Struc-
tures,” International Association for Bridge and Structural
Engineers Reports, V. 41, Zurich, Sept. 1982.
5-16. “Vibration Tests on Bircherweid Foot-Bridge,”
International Civil Engineering Monthly, V. II, No. 4,
1971/1972, pp. 189-192.
5-17. Matsumoto, Y., et al., “Dynamic Design of Foot
Bridges,” Proceedings, IABSE, 1978, pp. 17-78.
5-18. Csagoly, P., and Dorton, R. A., “Proposed Ontario
Bridge Design Load,” PR 186, Ministry of Transportation
and Communications, Downsview, Ontario, Canada.
5-19. Csagoly, P.; Campbell, T.I.; and Agarwal, A.C.,
“Bridge Vibration Study,” RR 181, Ministry of Transporta-
tion and Communications, Downsview, Ontario.
5-20. “Military Fixed Bridges,” TM5-312, Superintendent
of Documents, Washington, D.C.
5-21. Federal Aviation Agency, “Aircraft Data” AC
#150/5325-5, 1975, and “Airplane Airworthiness,” Advisory
Circular.
343R-66 MANUAL OF CONCRETE PRACTICE
Myrtle Creek Bridge in California used precast elements.
CHAPTER 6—PRELIMINARY DESIGN
6.1—Introduction
The time invested in preparing the preliminary design of a
bridge is perhaps the most critical period in the design process.
The designers must take time to properly and meticulously
consider all factors that influence the project. These factors
transcend purely utilitarian and first-cost considerations.
At the preliminary design stage, consideration should be
given to implement details that increase durability, reduce
maintenance and facilitate inspection. Very often it is very
difficult to incorporate these features in the final design if
they had not been considered initially. An excellent source of
information is the NCHRP Report 349 Maintenance Consid-
erations in Highway Design.
In the preliminary design, it is recommended to have an
interdisciplinary meeting where the points of view of the
structural designer, together with those of soils, construction,
maintenance, and inspection engineers, environmentalists,
geologists, and other specialists, are considered in an open
forum. By so doing, the final structure will be better and prob-
lems will be avoided during subsequent stages of design,
construction, and particularly the useful life of the structure.
Bridges are important structures in transportation systems,
and they should have a long service life. Bridges have an
impact on the social life, as well as the economy, of the
region and communities which they serve. Frequently,
bridges are very large structures that have a definite impact
on the landscape. Bridges have been included in the coat of
arms of many cities, recognizing the pride and identification
communities have with these man-made structures.
This chapter includes many of the general factors which
should be taken into consideration in the preliminary plan-
ning and design of bridges, as well as those that are specific
for concrete bridges. A discussion of typical structural,
esthetic, construction, operation, and maintenance aspects
are presented for commonly used types of concrete bridge
structures. The list is not intended to be complete. Therefore,
suggestions contained herein should be regarded as general
guidelines only.
Through preliminary design and approximate dimensional
proportioning based on theory, practice, experience, and
judgment, suitable geometries and dimensions may be estab-
lished. This information should be confirmed by preliminary
structural design computations. The same procedure should
be followed for several types of structures and configura-
tions. Preparation of scaled drawings of profile elevations of
trial bridges and their components is a good aid to gain a
sense of proportion and initial esthetic value. Preliminary
cost estimates should be prepared for several alternatives.
The environmental, economic, and esthetic factors can
then be evaluated to establish the basis for recommendations
to be submitted to the proper agencies for approval. Upon
receiving an approval, with or without further revisions, the
engineer can then perform the final design and produce a
complete set of construction contract documents.
Several factors which have a direct bearing on the decision-
making process should be reviewed. First, the high cost of
construction site labor has forced changes toward industrial-
ized construction methods. Methods and equipment for
 BRIDGE ANALYSIS AND DESIGN
transporting and erecting concrete elements weighing over
100 tons (90 tonnes) are now available. Second, sociological
demands have focused more attention on esthetic and envi-
ronmental considerations. Consequently, the bridge engineer
is faced with the necessity of identifying a growing list of
environmental constraints, along with corresponding trade-
offs in the design process. Third, technological progress has
led to mass production of high-performance precast and
prestressed concrete members with compressive strengths of
over 10,000 psi (69 MPa). In longer span bridges, it may be
economical to reduce superstructure dead loads using lighter
weight concretes, higher strengths, and composite sections.
6.2—Factors to be considered
In selecting a type of structure for a particular site, the
following factors should be evaluated (not necessarily listed
in order of priority):
6.2.1 Loads
6.2.1.1 Permanent loads (time invariant)
a. Dead loads (superstructure and substructure elements
and superimposed loads, including present or future
wearing courses and utilities carried by the structure)
b. Earth pressures, at rest, active or passive (on abut-
ments, piles located on sloping ground, and retaining
walls, including hydrostatic forces)
6.2.1.2 Transient loads (time variant)
a. Vehicular live load and its derivatives (static and
dynamic effects, centrifugal and braking forces), and
pedestrian loads. Where appropriate, loads imposed by
other transportation modes such as rail transit,
airplanes at airport bridges, heavy timber, and steel
haulers at industrial locations.
b. Wind loads (on structure and vehicles)
c. Force effects due to strains induced by ambient temper-
ature variations, temperature gradients, differential
creep, shrinkage, relaxation of prestressing steel, and
elastic shortening
d. Secondary prestress effects
e. Friction restraint in sliding bearings and shear resis-
tance in neoprene pads
f. Hydrodynamic loads (waves and currents)
6.2.1.3 Exceptional loads
a. Earthquake, intensity, and duration depending on seis-
micity of the location. Consideration should be given to
the potential for soil liquefaction
b. Ice pressure and forces associated with floods
c. Differential foundation settlement
d. Collision of vehicles on a primary supporting element
e. Construction loads, including heavy vehicles, lifting
cranes, and dynamic effects while lifting segments
f. Special permit vehicles
6.2.2 Geometry
a. Number of lanes and width of shoulders and sidewalks
to establish the total deck width
b. Horizontal and vertical clearances, as affected by
superelevations, structure depth, support widths, and
overhead utilities
343R-67
c. Structure type: span length, span-depth ratio, precast or
cast-in-place, etc.
d. Obstruction of sight for traffic under bridge
e. Detour or maintenance of traffic during construction
f. Access for major utilities carried by or passing under
the structure
g. Horizontal and vertical curvatures, and skew angle if
any
h. Alignment with approaches
i. Interference with aircraft corridors
j. Icing effects where bridge curvature may increase acci-
dent potential
k. Deck drainage
l. Bridge profile for wind load and hydrodynamic forces
m. Need for aircraft warning lights on high bridges
n. Need for navigation lights on bridges over navigable
water
o. Need for fenders and other protection from ships for
bridges over navigable water
p. Need for crash walls for bridges over railroads
6.2.3 Corrosion protection—Corrosion protection is a
factor that must be considered in all structures. With the
increase in use of deicing salts in the colder climates, corro-
sion has become a major problem in bridges, particularly in
the decks. Particular attention should be given also to corro-
sion protection in areas exposed to aggressive corrosive
conditions such as marine and industrial environments.
During preliminary design this fact should be taken into
consideration by providing enhanced drainage of the deck,
avoiding areas where debris could accumulate, and directing
runoff away from structural members.
The use of increased concrete cover, high-strength
concrete, epoxy-coated bars, chemical admixtures in the
concrete, and cathodic protection are some of the measures
that can add to the resistance of the structure to corrosion.
6.2.4 Esthetic considerations—Due to their importance in
the environment, bridges should be designed to blend with
and enhance it. One source of information is ACI MP-1,
Esthetics in Concrete Bridge Design.
Improving the appearance of a bridge can usually be done
with a modest increase in cost.
Factors to be considered:
a. Harmony with adjacent structures and the environ-
mental setting
b. Use of appropriate shapes and surface treatments befit-
ting the locality or setting.
c. Climatic factors that may have an impact on certain
types of bridges
d. Design of shapes and details that will assure good
appearance after years of service of the structure
6.2.5 Subsurface conditions
a. Bearing capacity of foundation: spread footings, piles,
or caissons
b. Uniform or differential foundation settlement
c. Settlement or longitudinal movement of an adjacent
embankment that affects the performance of the bridge
d. Stability of cuts and fills adjacent to the structure
e. Stability of the geological setting
343R-68 MANUAL OF CONCRETE PRACTICE
f. Corrosive soils
g. Water table level and anticipated fluctuations
6.2.6 First cost and ease of maintenance
a. Total time of construction
b. Construction details and constructibility: repetitive
elements and simplicity of connections
c. Protection of bearings, concrete, and reinforcing steel
against deleterious elements such as salts, acid rain,
sulfate exposure, and other chemical agents
d. Total cost including approach cuts and fills
e. Possibility of future widening
f. Adaptability for repairs, replacement of the deck or
bearings, or strengthening
g. Access for inspection
h. Availability of materials and construction expertise
and facilities: local versus imported
i. History of maintenance of similar existing structures
and their performance in severe floods or seismic
activity
6.2.7 Life-cycle cost—The total cost during the expected
life of the structure should be considered using economic
models, such as present worth analysis, when there are
significant differences in maintenance, operation costs, and
schedules for the bridge alternatives considered. Initial costs
do not properly reflect true cost differences unless there are
no significant cost differences in life-cycle costs after
construction. A range of effective interest rates should be
used to account for the difficulty in predicting long-term
interest rates.
6.2.8 Safety
a. Safety of construction crews and adjacent occupants
from utilities, traffic (rail, highway, waterway), and
falling debris
b. Minimizing traffic hazards in completed structure
c. Insuring redundancy of structural components by
providing alternative paths in lieu of single load paths
to insure against catastrophic failure. This is particu-
larly important in seismic regions. Bridge spans shall
be tied together to guarantee overall structural integrity
d. Detailing to enhance structural integrity
e. Providing proper warning signs, traffic signals, and
illumination on and below the bridge
f. Selection of structure type such that support elements
will not unnecessarily be exposed to collision damage.
When exposure to damage is unavoidable, elements
should be protected with barriers, earth berms, or by
other appropriate means
g. Safety barriers to prevent debris or snow from being
blown on tracks or roadways
h. Chain link fencing to inhibit pedestrians from throwing
objects onto traveled way (road or track) below
6.2.9 Waterway crossings—Special requirements
a. Bridge length, bridge under clearance and pier spacing
to accommodate flood discharge
b. Backwater effects upstream and on piers: ice jams and
debris dams
c. Scour potential and erosion of the abutments and
approach fills
d. Cofferdams
e. Pier protection
f. Navigational requirements and approvals
g. Maintenance of waterway traffic during construction
h. Stream flow and ice thrust loads
6.2.10 Rail and transit bridges—Special requirements
a. Extra overhead and horizontal clearances
b. Unusually heavy loadings such as unit trains, etc.
c. Operational restrictions and approvals during construction
d. Maintenance of rail traffic
e. Deck details and maintenance requirements for ballast
or direct fixation in transitways
f. Provision for future track raising
g. Provision for protection from exhaust emissions from
locomotives to the structure and users in the case of
underpasses
h. Security fencing provisions and barrier walls for
confinement purposes in case of derailments
i. Esthetic effort to coordinate these structures and design
them to be compatible with highway and pedestrian
bridges and the surroundings
6.2.11 Prestressed concrete—Special considerations
a. Cost of tooling (formwork, etc.)
b. Length of production run
c. Plan production versus on-site work
d. Weights and sizes of elements for transportation and
lifting
e. Transportation and erection equipment limitations and
costs
f. Material versus production cost
g. Availability of qualified personnel, materials, and
facilities
h. Construction sequencing and stage loading, particu-
larly with regard to post-tensioned girders, precast, and
cast-in-place segments
6.2.12 Environmental exposure
a. Chemical agents such as acid rain and salt
b. Industrial waste
c. General climatic factors such as relative humidity,
snow accumulation, and ice formation
d. Abrasive erosion on piers
6.3—High priority items
In general, after considering the preceding factors, the
following items should be finalized as early as practicable:
6.3.1 Typical section and alignment: Vertical and hori-
zontal—To avoid wasteful redesigns during the final stages,
it is essential to finalize, at the early design stages, bridge
parameters such as roadway alignment, number of lanes,
width of shoulders, type of curb and railing, waterway clear-
ances, and the number and size of utilities to be carried
across the structure. It is necessary that these items be shown
in the preliminary design drawings, often called TS&L (type,
size, and location) drawings, and submitted to all proper
agencies for approval at the preliminary design phase.
6.3.2 Span length composition: Uniform or varying—For
grade separation bridges, the total length is determined by
the requirements of the facility passing under the structure.
 BRIDGE ANALYSIS AND DESIGN 343R-69

Long spans that provide more than the minimum legal clear- require special falsework. Falsework bents adjacent to the
ances are desirable because of the increased safety, and are traffic should be protected.
generally more pleasant in appearance. In some bridges the In general, continuous structures with as few joints as
location and type of supports are dictated by foundation possible should be selected. Transverse joints should be
conditions, required clearances, safety, and esthetics. Long avoided if at all possible. Joints usually lead to higher initial
spans with few supports have pleasing appearances and costs and are the source of problems and higher maintenance
create the least obstruction. However, sometimes they have costs in bridges. Decreasing the number of joints also helps
a higher initial cost than shorter span structures dictated by to provide a smoother ride.
the minimum requirements. In longer bridges, expansion joints are unavoidable. The
An analysis of the span composition versus cost is spacing of joints depends on climatic conditions, as well as
appropriate. the type of structure. In milder climates, spacing of joints as
For river and valley crossings, the span lengths may be large as 300 ft (90 m) have been used successfully. Expan-
dictated by any combination of the following items: sion joints should be placed at points of discontinuity,
a. Foundation requirements usually on top of bents.
b. Geological setting and stability When the main load-carrying elements are cast in place,
c. Height of piers the deck slab is also cast in place. If the main elements are
d. Esthetics and harmony with the environment precast, the deck slab may consist of precast deck planks
e. Waterway area and other hydraulic considerations such with a cast-in-place topping. In this case, the planks and
as scour and high water levels during floods topping are generally designed and built for composite
f. Navigational requirements, both during construction action between the precast and cast-in-place portions.
and service Precast butted boxes, double tees, and decked bulb tees
g. Type of debris expected during flood stages (integral deck) may not need a cast-in-place concrete deck,
h. Reliability of flood magnitude predictions in particular, when laterally post-tensioned.
i. Ease of construction 6.4.1 Nonprestressed concrete slab bridges
j. Economy
k. Vessel collision requirements Span ranges
The depth, difficulty, and cost of constructing foundations Types Highway Railroad
determines span length after other constraints are applied. Solid slabs 16 to 44 ft 12 to 24 ft
Least overall first cost often results when substructure and (5 to 14 m) (4 to 8 m)
superstructure costs are equal. Cored or voided slabs 40 to 65 ft 20 to 40 ft
6.3.3 Special conditions—Special conditions include (12 to 20 m) (6 to 12 m)
special subsurface conditions, existing utilities, or mainte- Waffle soffit slabs 40 to 80 ft Not used
nance of traffic which might control the selection of the (12 to 24 m)
structure type, size, and location. Depth-to-span ratios
6.3.4 Combination with other structures—Combination Simple spans 1/15± 1/12±
with other structures such as dams and buildings. Continuous spans 1/20 to 1/24 1/16±
6.3.5 Environmental impact factors—Factors that have a
Appearance: Neat and simple; desirable for short spans.
significant environmental impact should be identified and
Construction: Simplest type for details and formwork.
taken care of during the preliminary design stage. Some of
Construction time: Shortest of any cast-in-place construction.
the factors to be considered are:
Maintenance: Very little except at transverse deck joints,
a. Wetlands
when used. Bearings usually require little maintenance.
b. Vegetation
6.4.2 Nonprestressed concrete girder bridges
c. Animal life
6.4.2.1 T-Beam (Fig. 6.4.1a)
d. Disruption to existing communities (human, flora,
fauna)
e. Integration with the setting Span ranges
f. Views Highway Railroad
30 to 120 ft (9 to 36.5 m) 20 to 40 ft (6 to 12 m)
6.4—Structure types Depth-to-span ratios
In the following subsections, the general characteristics of Simple spans 1/13± 1/8±
the common types of concrete bridge structures are given for Continuous spans 1/15± 1/10±
reference in preliminary studies. In Section 2.5 more infor- Appearance: Elevation is neat and simple. Bottom is
mation is given on this subject. Unless otherwise specifically cluttered. Horizontal widening of stems at piers is detracting. If
noted, the span-to-depth ratio and span lengths refer to required, the exterior girder should be designed with the
highway bridges only. Bridges carrying railroads generally widening on the inside face. Utilities, pipes, and conduits can
have smaller span-to-depth ratios and shorter spans. When- be concealed between girders.
ever traffic beneath the proposed bridge has to be maintained Construction: Requires a good finish on all surfaces; form-
during construction, the use of cast-in-place concrete may work may be complex.
343R-70 MANUAL OF CONCRETE PRACTICE

Appearance: Neat and clean lines from all views; utilities,

pipes, and conduits can be concealed between girders or
within boxes.
Construction: Rough form finish is satisfactory on inside
surfaces but removal of upper slab forming and supports is
difficult and the material may be left inside.
Construction time: More than required for slab or T-Beam
bridges due to staging of concrete placement.
Maintenance: Low, except that bearing and transverse
deck joint details may cause some problems.
6.4.3 Prestressed concrete slab bridges
6.4.3.1 Cast-in-place post-tensioned

Span ranges
Type Highway Railroad
Solid slabs Up to 100 ft Up to 20 ft
(30.5 m) (6 m)
Cored or voided slabs Up to 150 ft Up to 40 ft
(46 m) (12 m)
Partial box voided slabs (slabs Over 150 ft No data
which typically are solid over the (46 m) available
supports and a certain distance to
each side, but voided at the cen-
tral portion of the span)
Depth-to-span ratios
Simple 1/20± 1/16±
Continuous 1/24 to 1/40* 1/18±

Appearance: Neat and simple; desirable for relatively

short spans.
Construction: More complex than nonprestressed
concrete. Sequence of prestressing and grouting requires
supervision by specialists.
Construction time: Usually longer than as nonprestressed
concrete slabs, because of draping of conduit, placing of
anchors, attaining strength, and tensioning operations.
Maintenance: Very little except at transverse deck joints.
Maintenance is required at bearings and post-tensioning
embedments.
6.4.3.2 Precast pretensioned (Fig. 6.4.1f)
Fig. 6.4.1—Typical concrete bridge configurations.
Span ranges
Construction time: Usually somewhat longer than Types Highway Railroad
required for slab bridges due to more complex forming. Solid slabs 19 to 30 ft (6 to 9 m) 12 to 20 ft (4 to 6 m)
Maintenance: Low, except that bearing and transverse Cored slabs 30 to 80 ft (9 to 24 m) 24 to 40 ft (8 to 12 m)
deck joint details may require attention. At stream crossings, Depth-to-span ratio
with inadequate underclearance, floating debris may damage 1/25 to 1/33 1/20 ±
the girder stems. Appearance: Neat and simple; desirable for low short
6.4.2.2 Nonprestressed concrete box girder (Fig. 6.4.1.b) spans.
Construction: Details and formwork very simple; plant
Span ranges fabrication methods are suitable; field erection may be fast.
Highway Railroad No falsework required; units placed by cranes; no prolonged
80 to 200 ft (24 to 60 m) 50 to 100 ft (15 to 30 m) impediment to traffic.
Construction time: Time for erection of precast elements
Depth-to-span ratios at site is minimal.
Highway Railroad
Simple 1/14± 1/11± * For the larger span-to-depth ratios, a haunched slab is sometimes desirable.
Haunches should be designed using the longest possible curves to present the best
Continuous 1/18± 1/14± appearance
 BRIDGE ANALYSIS AND DESIGN 343R-71

Maintenance: Very little except at transverse deck joints,

bearings, longitudinal joints, and connections between units.
6.4.4 Prestressed concrete girder bridges
6.4.4.1 Cast-in-place, post-tensioned (Fig. 6.4.1a and
Fig. 6.4.1b)

Span ranges
Types Highway Railroad
Constant-depth beams 40 to 120 ft Up to 80 ft
(12 to 36 m) (24 m)
Constant-depth box beams 80 to 300 ft Up to 110 ft
(25 to 90 m) (30 m)
Haunched beams Up to 200 ft No data
(61 m) available
Haunched box beams Up to 700 ft No data
(215 m) available Fig. 6.4.4.2—Precast prestressed I-beam bridge.
Depth-to-span ratios
Simple spans 1/22± 1/14±
Continuous spans 1/25 to 1/33 1/16± Exterior girder should be designed with flat exterior faces
Haunched girders 1/33 to 1/50 No data or with exposed flanges running continuously over bents by
available widening web into end blocks on the interior face of the exterior
Appearance: From below, T-beams appear cluttered while girder, in post-tensioned beams.
a box girder is neat and clean, making it more attractive; Construction: Longer span girders are more difficult to
elevation of both types is neat and simple. Bent caps should handle after fabrication in plant and at job site; standard
be hidden between girders. Drop caps should be avoided for girders are preferred; special sections are usually more
best appearance. expensive; standard forms are stocked by fabricators;
Construction time: Somewhat longer than for nonpre- members can be used as simple spans, but preferably girders
stressed concrete T-beam or box girder bridges. should be made continuous for live load. Precast girders with
Maintenance: Very low except that bearing and transverse the deck slab cast-in-place can also be used. No falsework is
required except for long spans with precast segments post-
deck joint details require attention. Addition of transverse, as
tensioned; if the latter are stage-stressed by a cantilever
well as longitudinal post-tensioning, greatly reduces the
method, the falsework may not be needed.
number of cracks.
Very long T- and I-sections have shown substantial
6.4.4.2 Precast T-Beam, I-girder, and box girder camber and differential camber. Camber of bulb T-beams on
(Fig. 6.4.1.c, d, and e; and Fig. 6.4.4.2) curved bridges (each simple span being composed of straight
beams) has been reported as a major drawback.
Highway Railroad Construction time: Erection time for precast elements at
Span range site is minimal, but formwork may be required for slabs
30 to 160 ft Up to 110 ft between girders.
(9 to 50 m) (33 m)
Maintenance: Low except that bearing and transverse deck
Depth-to span ratios
joint details may require attention.
T-beam and I-girder
6.4.5 Post-tensioned segmental bridges—Bridges of the
Simple spans 1/18± 1/14±
various types may be constructed in segments and post-
Continuous spans 1/20± 1/16±
tensioned to complete the final structure. The segments may
Spread box girder
be cast-in-place or precast. If cast-in-place, it is common
Simple spans 1/18± Not used
practice to use the balanced cantilever construction method
Continuous spans 1/22± Not used
with traveling forms. If precast, the segments can be erected
Butted box girder
by any of four major methods: balanced cantilever, span-by-
Simple spans 1/25± 1/16±
span, incremental launching, and progressive placing. There
Continuous spans 1/28± 1/18±
are also numerous variations of these, depending on the
Precast box beams ordinarily have a width x of 3 to 4 ft erection techniques used.
(0.9 to 1.2 m), and heights y of 1 ft-9 in. to 4 ft (0.5 to 1.2 m). Segmental construction reduces or eliminates the need for
Appearance: I-beam is similar to T-beam, except for centering or falsework. It permits the use of longer precast
bulbous bottom of standard prestressed girder sections; members which can be transported from the precast plant to the
spread box beam is similar to T-beam; butted box girder is bridge site. It also requires much less time to construct at the
similar to cast-in-place box girder. Utilities, pipes, and site than a similar type cast-in-place. It permits construction in
conduits can be concealed between girders. a planned manufacturing cycle using repetitive operations.
343R-72 MANUAL OF CONCRETE PRACTICE
Fig. 6.4.6—Rigid frame bridge.
Fig. 6.4.8.1—Arch bridge.
Balanced cantilever construction should be carefully
designed where change in depth is required. Arch forms
joining each other or joining straight drop in spans require
esthetic consideration to produce the best appearance. Piers,
bents, or columns must also be recipients of esthetic consid-
erations, especially where they join the girders.
6.4.6 Rigid-frame bridges (Fig. 6.4.6)
Structural: Integral rigid negative-moment knees greatly
reduce the positive span moment and overturning moment at
foundation level; single rigid portal frames can be adapted to
narrow water channels, railways, subways, and divided or
undivided highways underneath; double-span rigid frames
are suitable for divided multilane highways underneath, with
sufficient median width; triple-span rigid frames can accom-
modate multilane divided highways with a wider center
median. The horizontal member may be of most of the
construction types described in Sections 6.4.1 through 6.4.4.
Members with variable moment of inertia are advantageous
and can be easily incorporated. Preliminary proportioning
can start with a thickness at the knee approximately equal to
twice that at the crown.
Appearance: Graceful and clean; well adaptable to stone
facing.
Construction: Simple formwork, but strong. Usually
requires curved formwork for variable depth.
Construction time: Similar to that of other cast-in-place types.
Maintenance: Low.
6.4.7 Jointless bridges—Movable joints in bridges are a
source of problems, and therefore, their number should be
minimized. Joints in the bridge deck are very costly to main-
tain, while detracting from the smoothness of traveling the
bridge. If maintenance is not adequate, water, deicing salts,
and other chemicals may leak downward, creating corrosion
problems in the supports if they are metallic, and areas where
deterioration of the concrete may start due to cycles of wet
and dry conditions and chemical action.
To avoid these problems there has been a tendency to
eliminate joints as much as possible, particularly in medium
length bridges. In many jurisdictions it is standard to build
jointless bridges up to 300 ft (90 m) long.
The substructure should be designed with appropriate
flexibility such as to maintain secondary stresses within
tolerable limits.
Jointless bridges require careful consideration of stresses
on the abutments and the design of the approach slab. It is
preferable to use integral abutments with the approach slab
made continuous with the bridge deck. An expansion joint is
placed between the approach slab and the road pavement.
6.4.8 Arch bridges
6.4.8.1 General (Fig. 6.4.8.1)
Structural: The arch structure is primarily a compression
member; constant depth for small spans and variable
moment of inertia for medium and long spans; spans as long
as 1000 ft (300 m) have been built; rise-to-span ratio varies
with topography, but the higher the rise, the less the dead
load thrust and temperature moment; for fixed arches, thick-
ness at springing lines usually is slightly more than twice that
at the crown; filled spandrels are used only with short spans;
for medium and long deck spans, open spandrels with road-
ways carried by columns are the rule; in a through arch, the
roadway deck is carried by hangers; long single spans are
used over deep waterways and shorter multiple spans are
more economical over wide shallow waters with excellent
foundation conditions. Shape could be curvilinear, usually
parabolic, or formed by straight segments whose vertexes
follow a parabola.
Appearance: Graceful and attractive, especially over deep
gorges, ravines, or large waterways.
Construction: Either using falsework or cantilever
methods. The cantilever method is preferred when circum-
stances prevent the construction of regular falsework or
make its construction too costly.
Construction time: Usually longer than for other types: use
of prefabricated segments and post-tensioning when shorter
time is desired.
Maintenance: Low.
6.4.8.2 Spandrel-filled arch (Fig. 6.4.8.2)
Structural: Suitable for short spans, 50 to 100 ft (15 to 30 m),
and low spandrels, particularly if live loads are heavy. For
longer spans, dead load stresses due to fill become excessive.
Usually its first cost is higher than other types of arches.
Used where esthetic considerations make it desirable.
6.4.8.3 Barrel arch (Fig. 6.4.8.3)
Structural: Suitable for spans longer than 100 ft (30 m); the
spandrel area is open and the deck supported on walls. Arch
rib is single unit with a width equal to or slightly less than the
deck width. Generally fixed, but may be two-hinged. Top of
arch rib at crown is just below the deck level.
6.4.8.4 Two-hinged rib open-spandrel arch (Fig. 6.4.8.4)
Structural: Usually used in spans longer than 300 ft (90 m).
The two hinges usually are placed at spring lines, either at
 BRIDGE ANALYSIS AND DESIGN
Fig. 6.4.8.2—Spandrel-filled arch.
Fig. 6.4.8.3—Barrel arch.
Fig. 6.4.8.4—Two hinged rib open spandrel arch.
Fig. 6.4.8.5—Fix rib open-spandrel.
the same level or not, depending on topography. Constant rib
section is generally used. Design of the hinges requires
special attention for heavy long-span arches. In deck-type
arch bridges, the roadway is supported above the arch by
columns carried by one or more arches; in through-type arch
bridges, the roadway is supported below and between the
arches by hangers. Ribs should be adequately braced trans-
versely by cross struts. Top of deck arch rib at crown is
generally well below deck level to permit use of columns.
6.4.8.5 Fixed-rib open-spandrel arch (Fig. 6.4.8.5)
Structural: Rib sections are generally variable; thinner
sections at crown reduce temperature effects, moments, and
dead load thrust; other factors the same as for two-hinged
arch. Usually used in spans over 300 ft (90 m).
6.4.8.6 Tied arch (Fig. 6.4.8.6)
Structural: Tied-arch bridges are used where the supporting
foundation cannot resist the arch thrust, or where clearance
requirements below the bridge restrict the depth of construction.
343R-73
Fig. 6.4.8.6—Tied arch.
Fig. 6.4.8.7—Stiffened arch.
Fig. 6.4.9—Cable-stayed bridge.
The horizontal thrust of the arch is resisted by the tie. The tied-
arch is always used as a through span. The deck floor of a tied-
arch bridge is always carried by hangers. Maintenance of ties
and hangers is critical due to tension cracking. Prestressed ties
and hangers should be considered to reduce maintenance.
Spans usually are longer than 300 ft (90 m).
6.4.8.7 Stiffened arch (Fig. 6.4.8.7)
Structural: Very similar to tied-arch bridge, with the
difference that the deck is provided with stiff longitudinal
girders. The arch and deck girders combined provide for a
more efficient system to resist the effect of moving live
loads. Same range of spans, over 300 ft (90 m).
6.4.9 Cable-stayed bridges (Fig. 6.4.9)
Structural: Stayed girder bridges have been used with
main span lengths as long as 1300 ft (396 m) (Dames Point
343R-74 MANUAL OF CONCRETE PRACTICE
Fig. 6.4.11a—Vierendeel truss.
Fig. 6.4.11b—Warren truss.
Fig. 6.4.11c—Pratt truss.
Bridge, Jacksonville, Fl.). This type of bridge is being used
in crossings where, previously, suspension bridges were the
preferred alternative. Concrete has been used in the substruc-
ture, and many times in the towers and deck. Cellular piers,
to carry towers and cable stays, have been used.
Towers have been built either solid or hollow, depending
on the size and loads involved. Concrete box girders have
been used extensively. They are either single or multiple
cells depending on the deck width. Particular attention must
be given during design to take care of stages of construction
and the different load conditions, which in many cases are
more critical than the final loads.
Protection of cable stays from fatigue and stress corrosion
has been a problem, especially at cable anchorages. High-
strength concrete has been used to reduce dead load (East
Huntington, W.V.).
Appearance: Graceful and delicate; well adapted to both
urban and rural environments.
Construction: At difficult crossings these bridges are
simpler to construct than other common bridge types;
stayed- type bridges are well adapted to long spans. False-
work usually is not necessary; deck may be constructed by
the balanced cantilever method, or precast. Asymmetric
structures have been successfully built at difficult locations.
Construction time: No more time than for other types of
similar span length.
Maintenance: Above average because of complexity of
suspension system.
6.4.10 Suspension bridges
Structural: Suspension bridges with concrete decks and
girders have been built in North America with main spans as
long as 680 ft (207 m) (Hudson Hope Bridge, Peace River,
B.C.). Main supporting elements are a pair of large steel
cables. Concrete deck system serves only to span between
hangers, to stiffen the structure and to resist horizontal and
live loads. To reduce dead load, high-strength and light-
weight concrete can be used for longer spans. Requires large
anchorages for suspension cables.
Appearance: Graceful, slender, well suited to long water
crossings.
Construction: Requires no falsework, but placement of
deck units should be properly staged. Lack of falsework
removes any impedance to traffic underneath during
construction.
Construction time: Probably somewhat greater than for a
cable-stayed bridge.
Maintenance: Above average due to generally greater
number of hangers.
6.4.11 Truss bridges (Fig. 6.4.11a, b, and c)
Structural: A truss bridge may be constructed as a simple,
cantilever, or continuous structure. It may have parallel
chords, inclined upper chord in a through type, or inclined
lower chord in a deck type. It may be prestressed, precast,
and field jointed, or precast and post-tensioned. For spans
longer than usual with girders, in the range of 120 to 250 ft
(36 to 76 m), it may prove competitive. A depth-span ratio of
1/10 or more should be used for a deck truss. The use of
inclined chords, particularly in the end panel, is recom-
mended. For about equal quantity of materials, a truss has
less deflection than its girder counterpart.
Although nonprestressed cast-in-place concrete truss
bridges of triangular configuration have been built, their use
is not recommended. The detail of reinforcement at a joint
where many members meet is very complex. Formwork,
centering, and placing of concrete is expensive, therefore
concrete truss bridges with structural steel diagonals offer a
solution to some of the problems mentioned.
Some precast pretensioned concrete trusses have been
built recently in which components are assembled into large
modules. The modules are erected by large launching trusses
and joined by external prestressing tendons.
While not used extensively, the Vierendeel truss offers
some esthetic qualities, has simpler details because of the
limited number of members at a joint, is easier to form and
place, and can be precast or cast in place. If necessary, it can
be erected by the cantilever method without falsework.
Appearance: Varies with shape, size, and design of the
joints. It could be the least attractive of any bridge type, but
if sensibly designed could be very interesting.
In general, the least number of parts produces the best
appearance. A Warren truss presents a better appearance
than a Pratt truss. A Vierendeel truss has the possibility of
presenting the best appearance. For greater strength and
rigidity in a Vierendeel truss, vertical members should be
spaced closer near the supports; however, the variation of
 BRIDGE ANALYSIS AND DESIGN 343R-75

spacing should be a minimum, as such variations complicate

deck details and increase cost. Effective span length* Slab thickness, h
Construction: Complicated forming slows the construc- ft m in. mm
tion. Requires extensive falsework, making it difficult to 6 1.8 71/4 185
continue traffic under the bridge unless cantilever method of 7 2.1 73/4 195
construction is used. 8 2.4 8 205
9 2.7 81/2 215
Construction time: Usually quite long.
10 3.0 83/4 220
Maintenance: Above average because of complexity of
11 3.4 9 230
trusses.
12 3.7 91/2 240
6.4.12 Special systems—This report should not be inter- 13 4.0 93/4 250
preted as having identified all concrete bridge types avail- 14 4.3 10 260
able for the application at a particular project. Special 15 4.6 101/2 270
systems using creative design and innovative construction 16 4.9 11 280
techniques are continually being developed and used. Each These recommended slab thicknesses will permit proper
concrete bridge project should be evaluated independently. placement of reinforcement with the required concrete
The following list of items is offered as a partial check list: protection. In areas of adverse climatic conditions, a thicker
a. Type of structure slab may be required. The designer should check the current
b. Span length practice of the highway authority in the area where the
c. Design loadings bridge is to be built.
d. Number and width of lanes For precast butted box-beams, a top slab thickness of 3 in.
e. Sidewalk(s) and bicycle way (75 mm) may be used.
f. Span-to-depth ratio For nonprestressed concrete girders, the stem width should
g. Substructure protection be at least 11 in. (280 mm) for up to eight longitudinal rein-
h. Strength and type of concrete forcing bars. It is often necessary to increase this thickness over
i. Grade of reinforcing steel continuous supports to accommodate compressive forces.
j. Type of prestressing system For post-tensioned girders, a 12-in. (300-mm) minimum
k. Method of fabrication and erection stem width provides for two rows of post-tensioning tendons.
l. Method of construction The bottom slab thickness of box girders should be approxi-
m. Site restrictions mately 1/16 of the clear span between webs, but should not
n. Environmental constraints exceed the thickness of the top slab, or be less than 5-1/2 in.
o. Initial cost and maintenance costs (140 mm). The bottom slab thickness may be increased near
The information in this chapter could also be used in the continuous supports to resist compressive forces.
preliminary design of special systems and unusual structures. The webs of box girder superstructures should have a
minimum thickness b of 8 in. (200 mm) unless external
6.5—Superstructure initial section proportioning prestressing is used. AASHTO recommends a minimum
The slab should be cantilevered from the exterior beam or thickness of 8 in. (200 mm) with no longitudinal or vertical
web of the box girder. This produces a more pleasant appear- post-tensioning tendons, 12 in. (305 mm) with only longitudinal
ance and is economical. Usually the cantilever length is from (or vertical) post-tensioning tendons, and 15 in. (381 mm) in
1/3 to 1/2 of the girder spacing. webs with both longitudinal and vertical tendons. It is often
Height of parapet, including slab thickness, should be beneficial to increase the thickness of webs near supports to
from 1/2 to 1/3 of the beam depth to provide a slender provide adequate shear resistance.
appearance to the beam. Minimum height to comply with For segmental precast segments, smaller dimensions than
safety requirements should be provided. those indicated may be adequate depending on size,
The girder spacing should be selected to accommodate the construction facilities available, etc.
deck width and to minimize the cost while maintaining a Other esthetic effects can be created by inclining the
pleasing appearance. In general, it is more economical to use parapet face inward to make it look bright, and inclining the
a thicker slab with a wider girder spacing and cantilevered webs of box girders to give a more slender appearance.
slabs 1/3 to 1/2 of the girder spacing. Usual girder spacings
are in the range of 6 to 9 ft (2 to 3 m) for nonprestressed T- 6.6—Abutments
beam bridges, 7 to 11 ft (2 to 3.5 m) for nonprestressed box 6.6.1 Types—As shown in Fig. 6.6.1, there are two basic
girder bridges, 8 to 16 ft (2.5 to 5 m) for precast prestressed types of abutments: open-end (stub) (Figs. 6.6.1a through
I-beam bridges, and 7 to 12 ft (2 to 3,5 m) for spread precast, 6.6.1c) and closed-end or retaining (deep) (Fig. 6.6.1d
prestressed box-beam bridges. through 6.6.1i). Each has several subtypes as follows:
For preliminary proportioning, the following slab thick-
* The effective span is measured as the clear distance between webs of T-beams
ness h may be used for AASHTO HS20 design loading: or box girders or the top flanges of I-beams.
343R-76 MANUAL OF CONCRETE PRACTICE
Fig 6.6.1—Abutment types.
6.6.1.1 Open-end abutments—These abutments are
located near the top of the end slope of the approaching
roadway embankment.
a. Diaphragm or integral; the wall of the abutment is
constructed monolithic with the superstructure
b. Seat type; there is a joint between the superstructure
and the abutment.
c. Spill-through; often used when an additional span may
be needed in the future. It is built prior to constructing
the approach embankment; thus, the approach roadway
embankment spills through the columns.
6.6.1.2 Closed-end abutments
a. Cantilever; designed as a retaining wall resisting earth
pressure through cantilever action. High abutments
may have counterforts.
b. Restrained; for short bridges, the superstructure may be
 BRIDGE ANALYSIS AND DESIGN 343R-77

used to support the abutments against earth pressure.

Where subsoil at the foundation level of the abutment
is soft, struts may be placed under the pavement,
extending from abutment to abutment.
c. Rigid frame; in these bridges the abutment is integral
with the superstructure.
d. Cellular or vaulted; for tall abutments the effect of the
earth pressure is reduced and the settlement of the
approach roadway minimized by use of this type. The
front wall is in effect a pier, but the short end span is
hidden by the sidewalls.
e. Gravity or semigravity; the earth pressure is resisted in
full or in part by the weight of the thick wall. It is gener- Fig. 6.7.1—Solid pier.
ally associated with bridges carrying railroad traffic.
f. Mechanically stabilized earth; the earth pressure is
resisted by precast concrete slab elements anchored by
thin metallic or nonmetallic strips embedded in the
embankment. The reaction of the end span is usually
supported by a footing resting on the embankment.
6.6.2 Abutment type selection—Open-end abutments are
usually more economical than closed-end abutments. Since
most of the approach embankment can be placed before
construction of the abutment, there is less potential settle-
ment of road approaches than for the higher backfilled
closed-end type. If used without side walls, open-end abut-
ments allow flexibility for future widening of the roadways Fig. 6.7.2—Pile bent.
being spanned, and they greatly reduce the roadside hazard
produced with a closed-end abutment with nominal hori-
zontal clearance.
Of the open-end abutments, the diaphragm type is
preferred over the seat type since it eliminates expansion
bearings and troublesome deck joints. The diaphragm type,
however, does not permit unlimited thermal or shrinkage
movements and should not be used for structures over 300 ft
(90 m) long, unless provision is made for movement by
expansion joints elsewhere. When selecting the abutment
type, consideration should be given to the drainage design.
Open-end abutments may not be suitable for river cross-
ings where cost of protection (rip-rap or other measures)
may offset the cost advantages of this abutment type. Fig. 6.7.3—Multi-column bent.
6.6.3 Bridge abutment approach slab—A smooth transi-
tion from the approach roadway to the bridge deck is
provided by using approach slabs which span over possible maximum height-thickness ratio of 15, but should not be less
backfill settlement. The slab also helps in controlling runoff than 12 in. (300 mm) thick. This is the minimum thickness
drainage from entering the backfill. that should be used with two planes of reinforcing steel.
When an adequately designed concrete approach slab 6.7.2 Pile bents (Fig. 6.7.2)—These are usually the most
supported at one end by the bridge is provided, usually live load economical and are applicable for slab spans, T-beams, and
surcharge is not considered in the design of the abutment. other types of superstructures. They are suitable for stream
crossings when debris is not a factor. The preferable
6.7—Piers and bents minimum width of cap is 2 ft-6 in. (0.75 m).
The term “pier” pertains to an intermediate substructure When drilled shafts are used, they may be extended as
unit between the abutments. A “bent” is a type of pier circular columns having similar appearance to multicolumn
consisting of one or more columns in the substructure unit, bents. They may be connected by a grade beam or small wall
with or without a cap. at or below ground level.
6.7.1 Solid piers (Fig. 6.7.1)—These are located in Pile bents are unsightly for bridges 20 ft (6 m) or more in
streams where debris, ice, or fast current are present; are height.
preferable for long spans, and may be supported on spread 6.7.3 Multicolumn bents (Fig. 6.7.3)—These generally
footings or pile foundations. A pier wall should have a support dry land structures. They may be supported on either
343R-78 MANUAL OF CONCRETE PRACTICE
Fig. 6.7.4—Single-column pier.
Fig. 6.7.5—Mushroom pier.
spread footings or pile foundations. Columns may be
circular, rectangular, or variable in section to provide
esthetic effect. The height to thickness ratio of uniform
section columns should be in the range of 12 to 15 for
appearance, as well as structural reasons. Unnecessarily
thick columns can produce high temperature and shrinkage
stresses. Where large column bents are located in streams,
debris walls of less thickness than the columns, extending
from the footing to an elevation above high water, are gener-
ally used.
6.7.4 Single-column piers (Fig. 6.7.4)—They can some-
times be used to avoid skewed bents. They have a special
advantage for viaducts over city streets where the location of
columns is restricted. Their use should be avoided in areas of
high seismicity.
6.7.5 Mushroom piers (Fig. 6.7.5)—They are similar to
single-column bents, but without a double-cantilever girder
cap. They resemble a column with an integral drop panel in
flat slab construction. A slab bridge is carried continuously
over such piers with the slab thickened over the piers, thus
forming a mushroom-like construction. They have been used
for spans intermediate between slab bridges and continuous
girder bridges.
6.7.6 Towers—For cable-stayed and suspension bridges,
the towers may be single or twin, fixed or hinged at the base.
Twin towers may have vertical or inclined legs. The details
of towers are discussed in Chapters 10 and 11.
6.8—Appurtenances and details
During the preliminary design, consideration should also
be given to the appurtenances and details that may be neces-
sary for the proper functioning and service of the structure.
To maintain unity in the design, and avoid problems when
the final plans are prepared, proposed solutions should be
incorporated into the preliminary plans submitted for
approval.
Among items to be considered are the following:
Utilities
Bicycle paths
Protection to separate different types of traffic (automo-
biles and trucks, bicycles, pedestrians)
Lighting
Observation areas in scenic bridges
Drainage
Noise abatement
6.9—Finishes
The following components are candidates for architectural
finish:
Exterior railing
Exterior edge of slab
Exterior girder
Column, bent or pier
Wing wall
Abutment face
Integral formed textures on these components up to 1-1/2 in.
(37.5 mm) thick can dramatically influence the appearance
and provide at minimal cost increase a way to unify parts of
a single bridge or coordinate several bridges, tunnels, and
other related roadway structures.
The architectural finish can be produced with form liners
or chemical retarder applied to the forms and washed or air
blasted upon removal of the forms. ACI 303R, “Guide to
Cast-in-Place Architectural Concrete Practice” provides
valuable information on architectural finishes.
 BRIDGE ANALYSIS AND DESIGN
Structural tests of box girder bridge (photo courtesy of Portland Cement Association).
CHAPTER 7—STRENGTH DESIGN
7.1—Introduction
The recommendations in this chapter are intended for
application to all structural elements of bridge structures,
except those designed as shells, pipe culverts, and other
unusual structures. They may also be applied to elements
subjected to special loading conditions during the stages of
fabrication and transportation of precast concrete and during
placement of cast-in-place concrete.
7.2—considerations for analysis, design, and review
7.2.1 General—All members of statically indeterminate
structures should be designed for the maximum effects of the
loads specified as determined by elastic analysis, except as
provided for in Section 7.2.4 and 7.2.5. Recommended loads
and load combinations are discussed in Chapter 5.
7.2.2 Stiffness—All assumptions, adopted for computing
the relative flexural and torsional stiffnesses of continuous
beams and rigid frame members, should be consistent
throughout the analysis. The moments of inertia to be used in
obtaining the relative stiffnesses of the various members can
be determined from either the uncracked concrete cross
section, neglecting the reinforcement, or from the transformed
cracked section as long as the same method is used throughout
the analysis of a continuous or rigid frame structure.
The effect of haunches should be considered both in deter-
mining bending moments and in designing members.
7.2.3 Span length—The span length of members, not built
integrally with their supports, should be considered as the
343R-79
clear span plus the depth of the member. However, the span
need not exceed the distance between centers of support.
In the analysis of continuous and rigid frame members,
center-to-center distances between supports should be used
for the determination of moments. Moments at faces of
support may be used for design of members built integrally
with supports.
7.2.4 Analysis—Various acceptable methods of analysis
are described in Section 10.3. All methods of analysis should
satisfy the conditions of equilibrium, displacement compati-
bility, and stability at all points in the structure, along with
all magnitudes of loading up to the ultimate. In addition, all
serviceability recommendations of Chapter 8 should be
satisfied.
7.2.5 Redistribution—Negative moments calculated by
elastic analysis at the supports of continuous nonprestressed
flexural members for any assumed loading arrangement can
be increased or decreased as follows (ACI 318)
ρ – ρ′
20 ⎛ 1 – --------------⎞ , percent (7-1)
⎝ ρb ⎠
where
ρ = ratio of tension reinforcement = As/bd
ρ′ = ratio of compression reinforcement = As′/bd
ρb = reinforcement producing balanced condition
343R-80
The modified negative moments should be used for calcu-
lating moments at sections within the spans. Redistribution
of negative moments should be made only when the section,
at which the moment is reduced, is so designed that ρ or ρ-
ρ′ is not greater than 0.50 ρb, where ρb is calculated as
follows (ACI 318)
0.85 β 1 f c ′ 87,000
ρ b = -----------------------
- --------------------------
fy 87,000 + f y
0.85 β 1 f c ′ 600
ρ b = -----------------------
- -------------------
fy 600 + f y
where
fc′ = specified compressive strength of concrete
fy = design yield strength of nonprestressed reinforcement
β1 = factor used to determine the stress block in ultimate
load analysis and design
Negative moments calculated by elastic analysis at the
supports of continuous prestressed flexural members,
where bonded reinforcement is provided at supports in
accordance with Section 9.11, can be increased or
decreased by not more than
(30 – 47c/d) percent
for any assumed gravity loading arrangement. In the
previous expression, c is the distance from the extreme
compression fiber to the neutral axis, and d is the distance
from the extreme compression fiber to the centroid of the
tension steel.7-1
The modified negative moments should be used for calcu-
lating moments at sections within spans for the same loading
arrangement. In no case should the negative moments be
increased or decreased by more than 20 percent.
Negative moments should not be redistributed where
fatigue of reinforcement is a governing factor. This condi-
tion can occur where moving live loads contributes a signif-
icant part of the stresses in the reinforcement.
7.2.6 Composite concrete construction
7.2.6.1 General considerations—The recommendations
of this section provide for the design of composite flexural
members consisting of concrete elements constructed in
separate placements, but so interconnected that the elements
respond to superimposed loads as a unit.
The entire composite member, or portions thereof, may be
used in resisting the shear and the bending moment. The
individual elements should be investigated for all critical
stages of loading.
If the specified strength, unit weight, or other properties of
the various components are different, the properties of the
individual components, or the most critical values, should be
used in design.
In calculating the flexural strength of a composite
member, no distinction should be made between shored and
unshored members. However, unshored construction leads
MANUAL OF CONCRETE PRACTICE
to higher stresses at service loads, and may pose a problem
where fatigue is a major consideration.
All elements should be designed to support all loads intro-
duced prior to the full development of the design strength of
the composite member. Reinforcement should be provided
as necessary to prevent separation of the components and to
control cracking.
7.2.6.2 Shoring—When used, shoring should not be
(7-2) removed until the supported elements have developed the
strength required to support the prevailing loads and to limit
deflections and cracking at the time of shore removal.
7.2.6.3 Vertical shear—When an entire composite
member is assumed to resist vertical shear, design should be in
accordance with the requirements of Section 7.3.7, as for a
monolithically cast member of the same cross-sectional shape.
Web reinforcement should be fully anchored into inter-
connected elements in accordance with Section 13.2.
Extended and anchored shear reinforcement may be
included as ties for horizontal shear.
7.2.6.4 Horizontal shear—In a composite member, full
transfer of horizontal shear forces should be assured at
contact surfaces of interconnected elements. Design for hori-
zontal shear should be in accordance with the recommenda-
tions of Section 7.3.15.
7.2.7 T-girder construction—In T-girder construction, the
(7-3) girder web and slab should be effectively bonded together.
Full transfer of shear forces should be assured at the interface
of the web and the slab.
The effective slab width used as a girder flange should not
exceed one-fourth of the girder span; its overhanging width on
either side of the web should not exceed six times the thickness
of the slab nor one-half the clear distance to the next girder.
For girders having a slab on one side only, the effective over-
hanging slab width used as a girder flange should not exceed
one-twelfth of the girder span, nor six times the thickness of the
slab, nor one-half the clear distance to the next girder.
For isolated T-girders, where the flange is used to provide
additional compression area, only that part of the flange adja-
cent to the girder web, with a thickness at least one-half the
width of the girder web, should be used as compression area.
Also, the total width of the flange used as compression area
should not exceed four times the width of the girder web.
Load distributing diaphragms should be placed between
the girders at span ends and within the spans at intervals not
exceeding 40 ft (12 m). Diaphragms may be omitted where
tests or structural analysis show adequate strength.
Diaphragms for curved girders should be given special
consideration.
7.2.8 Box girder construction
7.2.8.1 General—This section pertains to the design of
simple and continuous spans of single and multiple cell box
girder bridges of moderate span lengths (see Section 6.5 for
typical span lengths and depth-to-span ratios).
Box girders consist of girder webs and top and bottom
slabs. To insure full transfer of shear forces, the girder web
and top and bottom flanges should be effectively bonded
together at their interfaces.
 BRIDGE ANALYSIS AND DESIGN
For curved girder bridges, torsion should be considered, and
exterior girder shears should be increased to account for torsion.
7.2.8.2 Lateral distribution of loads for bending
moment—The live load bending moment for each interior
beam in a prestressed box beam superstructure should be
determined using the method given in Section 10.5.
7.2.8.3 Effective compression flange width—The effec-
tive width of slab used as a girder flange should not exceed
one-fourth of the girder span; the overhanging width used as
a flange on either side of the web should not exceed six times
the least thickness of the slab, nor one-half the clear distance
to the next web.
For webs having a slab on one side, only the effective
overhanging width of slab used as a girder flange should not
exceed one-twelfth of the girder span, nor six times the least
thickness of the slab, nor one-half the clear distance to the
next web.
7.2.8.4 Slab and web thickness—The thickness of the
top slab for highway bridges should be at least 6 in. (150 mm)
for nonprestressed construction and at least 5-1/2 in. (140 mm)
for prestressed construction.
For highway bridges, the thickness of the bottom slab
should be at least one-sixteenth of the clear span between
webs, or 5-1/2 in. (140 mm), whichever is greater. The
minimum thickness may be reduced to 5 in. (127 mm) for
factory-produced precast elements.
The thickness of the bottom slab need not be greater than
the top slab, unless required by design.
If required for shear, the web may be thickened in the area
adjacent to the supports. The change in web thickness should
be tapered over a minimum distance equal to twelve times
the difference in web thickness.
For post-tensioned box girders, in order to accommodate
the post-tensioning ducts, the webs should be at least 1.0 ft
(300 mm).
The designer should note, however, that these minimum
thickness recommendations may not be adequate for heavily
reinforced members. For the top and bottom slabs in partic-
ular, where more than three layers of reinforcing are provided,
the thickness should be sufficient to provide adequate clear
cover, construction tolerances, and design depth.
7.2.8.5 Top and bottom slab reinforcement—Uniformly
distributed reinforcement of at least 0.4 percent of the flange
area should be placed in the bottom slab parallel to the
girder span. A single layer of reinforcement may be
provided. The spacing of such reinforcement should not
exceed 18 in. (500 mm).
Minimum distributed reinforcement of 0.5 percent of the
cross-sectional area of the slab, based on the least slab thick-
ness, should be placed in the bottom slab transverse to the
girder span. Such reinforcement should be distributed over
both surfaces with a maximum spacing of 18 in. (500 mm).
All transverse reinforcement in the bottom slab should
extend to the exterior face of the outside web in each group
and be anchored by a standard 90 deg hook.
At least one-third of the bottom layer of the transverse
reinforcement in the top slab should extend to the exterior
face of the outside web in each group and be anchored by a
343R-81
standard 90 deg hook. If the slab extends beyond the last
girder web, such reinforcement should extend into the slab
overhang and have an anchorage beyond the exterior face of
the web not less than that provided by a standard hook.
7.2.8.6 Diaphragms—Load distributing diaphragms or
spreaders should be placed at 60 ft (18 m) intervals, maximum,
unless tests or structural analysis show adequate strength. In
addition, diaphragms should be placed at main supports to
provide transfer of transverse wind loads to the substructure.
On curved box girders the need for diaphragms and spacing
requirements should be given special consideration.
7.2.9 Limiting dimensions for members
7.2.9.1 General—Because of the difficulty in placing
concrete and the increase in maintenance costs, the use of
thin or small members is seldom economically justifiable.
The designer should exercise good judgment in choosing the
optimum size of member.
7.2.9.2 Compression members—Circular compression
members, constituting the principal supports of a structure,
should have a diameter of at least 12 in. (300 mm). Rectan-
gular compression members should have a thickness of at
least 10 in. (250 mm) and a gross area not less than 100 in.2
(62500 mm2). Auxiliary supports should be not less than 6
in. (150 mm) minimum dimension.
7.2.9.3 Flexural members—The width of the compres-
sion face of flexural members should not be less than 6 in.
(150 mm). Structural slabs, including the flanges of T-
girders, should not be less than 4 in. (102 mm) thick;
however, in many situations, especially where more than two
layers of reinforcing are required, a greater thickness may be
needed to meet minimum cover recommendations of Section
13.8, and to provide required design depth.
7.3—Strength requirements
7.3.1 Required strength—Bridge structures and structural
members should be designed to have strength at all sections
sufficient to safely resist the structural effects of the load
groups which represent various combinations of loads and
forces to which the structure may be subjected, as stipulated
in Section 5.12. Each part of such structures should be
proportioned for the group loads that are applicable, and the
maximum design required should be used. The serviceability
requirements of Chapter 8 should also be satisfied to insure
adequate performance at service load levels.
7.3.2 Strength—The design strength provided by a
member or cross-section in terms of load, moment, shear, or
stress should be taken as the nominal strength calculated in
accordance with the recommendations and assumptions of
this Section 7.3, multiplied by a strength reduction factor φ.
Strength reduction factor φ should be as follows (ACI 318):
Flexural, without axial load 0.90
Axial tension, and axial tension with flexure 0.90
Axial compression, and axial compression with flexure:
Members with spiral reinforcement conforming
to Section 13.3 0.75
Members with ties conforming to Section 13.3 0.70
Except for low values of axial load, φ may be increased in
accordance with the following:
343R-82 MANUAL OF CONCRETE PRACTICE

For members in which fy does not exceed 60,000 psi (410 a. The concrete stress of 0.85fc′ is assumed, uniformly
MPa) with symmetric reinforcement, and with (h - d′ – ds)/h distributed over an equivalent compression zone,
not less than 0.70, φ may be increased linearly to 0.90 as φPn bounded by the edges of the cross section and a straight
decreases from 0.10fc′Ag to zero. line located parallel to the neutral axis at a distance a =
For other reinforced members, φ may be increased linearly β1c from the fiber of maximum compressive strain.
to 0.90, as φPn decreases from 0.10fc′Ag or φPb, whichever is b. Distance c from the fiber of maximum strain to the
smaller, to zero. neutral axis is measured in a direction perpendicular to
Shear and torsion 0.85 that axis.
Bearing on concrete 0.70 c. Factor β1 is taken as 0.85 for concrete strengths fc′, up to
Flexure in plain concrete 0.65 and including 4000 psi (27.6 MPa). For strengths above
For prestressed members produced in plants meeting the 4000 psi (27.6 MPa), β1 is reduced at a rate of 0.05 for
requirements of PCI Manual MNL-116, the following each 1000 psi (6.89 MPa) of strength in excess of
strength reduction factors should be used (AASHTO Stan- 4000 psi (27.6 MPa), but β1 need not be less than 0.65.
dard Specifications for Highway Bridges and AREA Manual 7.3.4 Flexure
for Railway Engineering, Chapter 8): 7.3.4.1 Minimum reinforcement of nonprestressed flex-
Flexure, with or without axial tension and for ural members—At any section of a flexural member where
axial tension 0.95 tension reinforcement is required by analysis, the reinforce-
Shear and torsion 0.90 ment provided should be adequate to develop a factored
Compression members, with prestress exceeding moment at least 1.2 times the cracking moment calculated on
225 psi (1.55 MPa) and with spiral reinforcement the basis of the modulus of rupture for normal weight
conforming to Section 13.3.2 0.80 concrete. The modulus of rupture should be obtained from
Compression members, with prestress exceeding tests, or may be taken as
225 psi (1.55 MPa) without spiral reinforcement 0.75
Bearing on concrete 0.75 7.5 f c ′ ( 0.623 f c ′ ) (ACI 318)
For all other prestressed members not specifically
covered, the factors for nonprestressed concrete should be The previous recommendations may be waived if the area
used. Development lengths specified in Chapter 13 do not of reinforcement provided at a section is at least one-third
require a φ factor. greater than that required by analysis based on the critical
7.3.3 Design assumptions—The strength design of members loading combinations.
for flexural and axial loads should be based on assumptions As an aid to the designer, the minimum recommended
given in this section, and on satisfaction of the applicable condi- reinforcement ratio ρmin7-2 may be obtained from the
tions of equilibrium and compatibility of strains. following approximate expressions
Strain in reinforcement and concrete should be assumed
directly proportional to the distance from the neutral axis,
( I ⁄ yt ) ( I ⁄ yt ) fc ′
except that, for deep flexural members with overall depth-to- ρ min = 10 + --------------
- --------------
- --------- (7-4)
2 2 fy
clear span ratios greater than 2/5 for continuous spans and bd bd
4/5 for simple spans, a nonlinear distribution of strain should
be considered (ACI 318).
Maximum usable strain at the extreme concrete compres- ⎧ ( I ⁄ yt ) ( I ⁄ yt ) fc ′ ⎫
ρmin = ⎨ 0.83 10 + --------------
- --------------- --------- ⎬
sion fiber should be assumed equal to 0.003, excluding 2 2 fy ⎭
⎩ bd bd
shrinkage, creep, and temperature strains (ACI 318).
Stress in reinforcement, below specified yield strength fy
for the grade of reinforcement used, should be taken as Es ( I ⁄ yt ) fc ′
ρ min = 10.2 --------------
- --------- (7-5)
times steel strain. For strains greater than that corresponding 2 fy
bd
to fy, stress in reinforcement should be considered indepen-
dent of strain and equal to fy (ACI 318).
Tensile strength of concrete should be neglected in axial ( I ⁄ yt ) fc ′
tension strength calculations and in flexural tension strength ρ min = 0.847 --------------
- ---------
2 fy
calculations of reinforced concrete, except when meeting the bd
requirements of Section 8.7.
The relationship between the concrete compressive stress fc ′
2
distribution and the concrete strain may be assumed to be ρ min = 1.7 ( h ⁄ d ) ---------
rectangular, trapezoidal, parabolic, or any other shape that fy
results in prediction of strength in substantial agreement with
the results of comprehensive tests. These recommendations fc ′
2
may be considered satisfied by an equivalent rectangular ρ min = 0.141 ( h ⁄ d ) --------- (7-6)
fy
concrete stress distribution defined by the following (ACI 318):
 BRIDGE ANALYSIS AND DESIGN 343R-83

Of the previous expressions, Eq. (7-4) is the most accurate where

and can be used for T-beams, box girders, and rectangular
sections. Eq. (7-5) is a somewhat simpler expression that can 0.85f c ′ ( b – b w )h f
be used for box girders or rectangular sections. The simplest A sf = ----------------------------------------- (7-11)
fy
expression, Eq. (7-6), can be used for the special case of rect-
angular sections.
and
These minimum reinforcement recommendations should
be followed, even where analysis shows that the calculated
( A s – A sf )f y
moment would be resisted solely by the tensile strength of a = ---------------------------
- (7-12)
the concrete. 0.85f c ′cb w
These minimum reinforcement recommendations do not
apply to footings. The balanced reinforcement ratio for flanged sections with
7.3.4.2 Maximum reinforcement of nonprestressed flex- tension reinforcement only can be computed as follows (ACI
ural members—For flexural members, the reinforcement 318R)
ratio ρ provided should not exceed 0.75 of that ratio ρb,
b 0.85β 1 f c ′ 87, 000 A sf ⎞
which would produce balanced strain conditions for the ρ b = -----w- ⎛ ---------------------
- ---------------------------- + --------
-
section under flexure (ACI 318). b ⎝ fy 87, 000 + f y b w d⎠
Balanced strain conditions exist at a cross section when
the tension reinforcement reaches its yield strength fy, just as
b 0.85β 1 f c ′ 600 A sf
the concrete in compression reaches its ultimate strain of ρ b = -----w- ⎛⎝ ---------------------- ------------------- + ---------⎞⎠ (7-13)
0.003 (ACI 318). b fy 600 + f y b w d
7.3.4.3 Rectangular sections with nonprestressed tension
reinforcement only—For rectangular sections, the design For T-girder and box-girder construction the width of the
moment strength can be computed as follows (ACI 318R) compression face b should be equal to the effective slab
width as defined in Section 7.2.7.
Mn = As fy d (1 – 0.59ρfy /fc′) 7.3.4.5 Rectangular sections with compression rein-
forcement—For rectangular sections and flanged sections in
which the neutral axis lies within the flange, the design
Mn = As fy ⎛ d – ---⎞
a
(7-7)
⎝ 2⎠ moment strength can be computed as follows (ACI 318R)

M n = ( A s – A s ′ )f y ⎛ d – ---⎞ + A s ′f y ( d – d′ )
where a
(7-14)
⎝ 2⎠
a = As fy /(0.85 fc′cb) (7-8)
where
The balanced reinforcement ratio for rectangular sections
( A s – A s ′ )f y
with tension reinforcement only may be calculated as a = ---------------------------
- (7-15)
follows (ACI 318R) 0.85f c ′b

0.85 β 1 f c ′ 87, 000 and the following condition should be satisfied

ρ b = -----------------------
- ----------------------------
fy 87, 000 + f y
As – As ′ f c ′d′ 87, 000
- ≥ 0.85β 1 ----------
------------------ ---------------------------
bd f y d 87, 000 – f y
⎛ ρ = 0.85 β 1 f c ′ 600 ⎞
-----------------------
- ------------------- (7-9)
⎝ b
fy 600 + f y⎠
( As – As ′ ) f c ′d′ 600
----------------------- ≥ 0.85β 1 ---------- ------------------ (7-16)
bd f y d 600 – f y
7.3.4.4 Flanged sections with tension reinforcement
only—When the compression flange thickness is equal to or
greater than the depth of the equivalent rectangular stress When the value of (As – As′)/bd is less than the value given
block, the design moment strength Mn may be computed by by Eq. (7-17), so that the stress in the compression reinforce-
the equations given in Section 7.3.4.3. When the compres- ment is less than the yield strength, or when the effects of
sion flange thickness is less than a, the design moment compression reinforcement are neglected, the design
strength can be computed as follows (ACI 318R) moment strength can be computed by the equations in
Section 7.3.4.3. In these cases, the section is treated as if
reinforced with tension steel only. Alternatively, a general
M n = ( A s – A sf )f y ⎛ d – ---⎞ + A sf f y ( d – 0.5h f )
a analysis can be made based on stress and strain compatibility
(7-10)
⎝ 2⎠ using the assumptions given in Section 7.3.3.
343R-84 MANUAL OF CONCRETE PRACTICE

The balanced reinforcement ratio for rectangular sections in which case

with compression reinforcement can be calculated as follows
(ACI 318R) Tf = 0.85fc′(b – bw)hf (7-23)

0.85β 1 f c ′ 87,000 f sb ′ In the previous expressions, dp is the effective depth of the

ρ b = --------------------- - ρ′
- -------------------------- + ------- prestressing steel. Design examples are given in Reference 7-3.
fy 87,000 + f y f y
b. For members with unbonded prestressing tendons and
a span-to-depth ratio of 35 or less (ACI 318)
⎛ ρ = 0.85β 1 f c ′ 600
---------------------
f sb ′ ⎞
- ρ′
- ------------------- + ------- (7-17)
⎝ b fy 600 + f y f y ⎠ fc ′
f ps = f se + 10,000 + --------------
-
100ρ p
where fsb′ = stress in compression reinforcement at balanced
conditions
fc ′ ⎞
⎛ f = f + 69 + --------------
- (7-24)
⎝ ps se
100ρ ⎠
d′ 87, 000 + f
p
f sb ′ = 87, 000 ⎛ 1 – ---- ----------------------------y⎞ ≥ f y
⎝ d 87, 000 ⎠
but fps in Eq. (7-24) should not be greater than fpy or (fse +
60,000) [(fse + 410)].
d′ 600 + f c. For members with unbonded prestressing tendons and
f sb ′ = 600 ⎛ 1 – ---- -------------------y⎞ ≥ f y (7-18)
⎝ d 600 ⎠ with a span-to-depth ratio greater than 35 (ACI 318)

fc ′
7.3.4.6 Other nonprestressed cross sections—For other - + 10,000
f ps = f se + --------------
cross sections and for conditions of nonsymmetrical bending, 300ρ p
the design moment strength φMn should be computed by a
general analysis based on stress and strain compatibility, using fc ′
⎛ f = f + --------------
- + 69⎞ (7-25)
the assumptions given in Section 7.3.3. The recommendations ⎝ ps se
300ρ ⎠
p
of Section 7.3.4.2 should also be satisfied.
7.3.4.7 Prestressed concrete members—The design but fps in Eq. (7-25) should not be greater than fpy , or (fse +
moment strength for prestressed flexural members can be 30,000) [(fse + 205)].
computed by the same strength design procedures and equa- Nonprestressed reinforcement conforming to Section 3.2,
tions recommended for nonprestressed members. For when used in combination with prestressing tendons, may be
prestressing tendons, fps should be substituted for fy. considered to contribute to the tensile force and may be
In lieu of a more accurate determination of fps based on included in the moment strength calculations at a strength
strain compatibility, and provided that fse is not less than equal to the specified yield strength fy.
0.5fpu, the following approximate values should be used:7-3 The index of prestressed and nonprestressed reinforce-
a. For members with bonded prestressing tendons,7-3 the ment used for the computation of the moment strength of a
equation fps = fpu(1 – 0.3 c/d) can be approximated by member should be such that c/du ≤ 0.42.7-3 For flanged
sections, the steel area should be required to develop the
fps = fpu (1 – 0.3c/du) (7-19) compressive strength of the web only.
When a reinforcement index in excess of that previously
where recommended is provided, design moment strength should
be based on the compression portion of the internal moment
resisting couple. The following expressions satisfy the intent
A ps f pu d p + A s f y d of this recommendation:
d µ = ------------------------------------------
- (7-20)
A ps f pu + A s f y a. For rectangular or flanged sections in which the neutral
axis is within the flange (ACI 318R)
and
2 2
M n = f c ′bd p ( 0.36β 1 – 0.08β 1 ) (7-26)
A ps f pu + A s f y – A s ′f y
c = ----------------------------------------------------------------- for β1c ≤ hf (7-21)
0.85β 1 f c ′b + 0.3A ps f pu ⁄ d u
b. For flanged sections in which the neutral axis falls
outside the flange (ACI 318R)
or
2 2
Mn = [ f c ′b w d p ( 0.36β 1 – 0.08β 1 ) +
A ps f pu + A s f y – A s ′f y – T f
- for β1c ≥ hf
c = -------------------------------------------------------------------- (7-22)
0.85β 1 f c ′b w + 0.3A ps f pu ⁄ d u 0.85fc′(b – bw)hf (dp – 0.5hf)] (7-27)
 BRIDGE ANALYSIS AND DESIGN
The total amount of prestressed and nonprestressed rein-
forcement should be adequate to develop a factored load at
least 1.2 times the cracking load, computed on the basis of
the modulus of rupture fr.
7.3.4.8 Special recommendations for slabs—The
minimum area of flexural reinforcement for one-way,
nonprestressed slabs in the direction of the span should be as
recommended in Section 7.3.4.1. The maximum area of flex-
ural reinforcement for one-way, nonprestressed slabs in the
direction of the span should be as recommended in Section
7.3.4.2. The design moment strength φMn of one-way,
nonprestressed slabs may be computed as recommended in
Section 7.3.4.3. The design moment strength φMn of one-
way, prestressed slabs may be computed as recommended in
Section 7.3.4.7.
The minimum area of shrinkage and temperature rein-
forcement for one-way slabs transverse to the direction of
the span should be as follows:
Slabs where Grade 40 or 50 deformed bars
are used 0.0020
Slabs where Grade 60 deformed bars are used 0.0018
7.3.5 Nonprestressed compression members with or
without flexure
7.3.5.1 General requirements—The design of members
subject to combined flexure and axial load should be based
on stress and strain compatibility using the assumptions
given in Section 7.3.3. For prestressed members, refer to
Chapter 9. Slenderness effects should be evaluated following
the recommendations of Section 7.3.6.
Members subject to compression load only, or to
combined axial load and flexural load, should be designed
according to the recommendations of Section 7.3.5.3.
7.3.5.2 Limits for reinforcement of compression
members—The area of longitudinal reinforcement for
compression members should not be less than 0.01 or more
than 0.08 times the gross area of Ag of the section.
The minimum number of longitudinal reinforcing bars in
compression members should be four for bars within rectan-
gular ties six for bars within circular ties, and six for bars
enclosed by spirals conforming to Section 13.3.
For compression members with a larger cross section than
required by considerations of loading, a reduced effective
area Ag may be used to determine the minimum longitudinal
reinforcing. In no case, however, should this effective area
be taken as less than one-half the total area.
7.3.5.3 Compression member strength—The following
provisions may be used as a guide to define the range of the
load-moment interaction relationship for members subject to
axial load only or combined axial load and flexural load.
The design axial load strength φPn of compression
members with spiral reinforcement as recommended in
Section 13.3 should not be greater than (ACI 318)
φPn = 0.85φ[0.85fc′(Ag – Ast) + fy Ast] (7-28)
The design axial load strength φPn of compression
members with tie reinforcement as recommended in Section
13.3 should not be greater than (ACI 318)
343R-85
φPn = 0.80φ [0.85fc′(Ag – Ast) + fy Ast] (7-29)
All members subjected to a compression load should be
designed for the maximum effects of factored moments and
axial loads. The factored axial load Pu at the given eccentricity
should not exceed φPn as calculated in Eq. (7-28) or (7-29).
The balanced strain conditions for a cross section are
defined in Section 7.3.4.2. For a rectangular section with
reinforcement in one or two faces and located at approxi-
mately the same distance from the axis of bending, the
balanced load strength φP and balanced moment strength
φMn can be computed as follows (ACI 318R)
φPb = φ(0.85fc′bab + As′fsb′ – As fy) (7-30)
and
φMb = [0.85fc′bab(h/2 – ab/2) + As′fsb′(h/2 – d′)
+ As fy(h/2 – ds)] (7-31)
where
87,000 ⎞ d
a b = ⎛ -------------------------- β
⎝ 87,000 + f ⎠ 1
y
600
a b = ⎛ -------------------⎞ β 1 d (7-32)
⎝ 600 + f ⎠
y
and
fsb′ = 87,000 – (d′/d)(87,000 + fy ) ≤ fy
[fsb′ = 600 – (d′/d)(600 + fy ) ≤ fy] (7-33)
The design strength under combined flexure and axial load
should be based on stress and strain compatibility using the
assumptions given in Section 7.3.3. The strength of a cross
section is controlled by tension when the nominal axial load
strength Pn is less than Pb. The strength of a cross section is
controlled by compression when the axial load design strength
Pn is greater than Pb. The combined axial load and moment
strength should be multiplied by the appropriate capacity
factor reduction φ as recommended in Section 7.3.2.
For members subject to combined flexure and compressive
axial load, when the design axial load strength φPn is less
than the smaller of 0.10fc′Ag or φPb, the ratio of reinforce-
ment ρ provided should not exceed 0.75 of the ratio ρb that
would produce balanced strain conditions for the section
under flexure without axial load. For members with compres-
sion reinforcement, that portion of ρb, equalized by compres-
sion reinforcement, need not be reduced by the 0.75 factor.
7.3.5.4 Biaxial loading—In lieu of a general section
analysis based on stress and strain compatibility for a condi-
tion of biaxial flexural loading, the design strength of rectan-
gular members under such loading conditions may be
343R-86
approximated using the charts, tables, and formulas given in
Reference 7-4, or the following expressions can be used to
approximate the axial design strength of noncircular
members subject to biaxial bending7-5
1 or beam, and is at least one-half the width of the
φP nxy = -------------------------------------------------------------------
1 ⁄ φP nx + 1 ⁄ φP ny + 1 ⁄ φP o
when
Pu ≥ 0.10fc′Ag
or
Mux /φMnx + Muy /φMny = 1
when the applied axial design load
Pu < 0.10fc′Ag
7.3.6 Slenderness effects in compression members
7.3.6.1 General—Wherever possible, the design of
compression members should be based on a comprehensive
analysis of the structure. Such analysis should be a second-
order analysis, taking into account the deformation of the
structure and the duration of the loads. When axial loads are
of sufficient magnitude to reduce stiffness or increase fixed-
end moments, such effects should be included.
In lieu of a second-order analysis, the design of compres-
sion members can be based on the approximate procedure
recommended in the following sections.
7.3.6.2 Unsupported length—For purposes of deter-
mining the limiting dimensions of compression members,
the unsupported length lu should be the clear distance
between lateral supports, except as recommended in Subsec-
tions (a) through (e) below:
a. In pile bent construction, lu for the pile should be the
clear distance between the lowest lateral support, as
described in Subsection (c) below, and the lower
extremity at which the pile may be assumed to be fixed,
depending on the soil conditions.
b. For compression members supported on spread or pile
footings, lu should be the clear distance between the top
of the footing and the underside of the deepest flexural
member, framing into the compression member in the
direction of potential translation at the next higher
level.
c. For members restrained laterally by struts, ties, or
beams, lu should be the clear distance between consec-
utive struts in the vertical plane, provided that two such
struts should meet the compression member at approx-
imately the same level, and the angle between vertical
planes through the struts should not vary more than
15 deg from a right angle. Such struts should be of
adequate dimensions and have sufficient anchorage to
restrain the member against lateral deflection.
MANUAL OF CONCRETE PRACTICE
d. For compression members restrained laterally by struts
or beams with brackets used at the junction, lu should
be the clear distance between the lower support or
fixity of the base and the lower edge of the bracket,
provided that the bracket width equals that of the strut
(7-34) compression member.
e. Where haunches are present, lu should be measured to the
lower extremity of the haunch in the plane considered.
The length which produces the greatest ratio of length to the
radius of gyration of the section should be considered, as well
(7-35)
as compatibility with the loading and support conditions.
7.3.6.3 Radius of gyration—The radius of gyration r
may be equal to 0.30 times the overall dimension in the
direction in which stability is being considered for rectan-
(7-36) gular members, and 0.25 times the diameter for circular
members. For other shapes, r may be computed for the gross
concrete section.
7.3.6.4 Effective length factor and lateral stability—The
(7-37) effective length of a compression member should be Klu,
where the effective length factor K should not be less than
(ACI 318)
a. For compression members braced against side-sway,
the effective length factor K should be 1.0, unless anal-
ysis shows that a lower value may be used.
b. For compression members not braced against side-
sway, lateral instability should be considered in deter-
mining the effective length Klu for all structures in
which the total translational stiffness of the bracing
elements is less than six times the sum of the combined
translational stiffnesses of all the compression
members in the level under consideration. All other
structures may be assumed to be restrained against
lateral instability. For structures subject to side-sway,
the value of Klu should be determined with due consid-
eration of cracking and reinforcement on relative stiff-
ness. When determining the stiffness ratio of beams
and columns to obtain K, the beam stiffness should be
obtained using Eq. (7-39), with ßd as zero. In no case
should K be less than 1.0.
For compression members braced against side-sway, the
effects of slenderness may be neglected when Klu/r is less
than 34 – 12M1/M2. For compression members not braced
against side-sway, the effects of slenderness may be
neglected when Klu/r is less than 22. For all compression
members with Klu/r greater than 100, an analysis should be
made as defined in Section 7.3.6.1.
For evaluation of slenderness effects, M1 is defined as the
smaller factored end moment, calculated by conventional
frame analysis. M1 is positive if the member is bent in single
curvature and negative if bent in double curvature. M2 is
defined as the larger factored end moment, calculated by
conventional frame analysis, always positive.
7.3.6.5 Moment magnification—Compression members
should be designed using the factored axial load Pu from a
conventional frame analysis and a magnified factored
moment Mc defined as follows (ACI 318)
 BRIDGE ANALYSIS AND DESIGN
Mc = δbM2b + δsM2s
where
Cm
δ b = ---------------------------
- ≥ 1.0
1 – P u ⁄ φP c
1 comprise a bent, or when they are connected integrally to the
δ s = ------------------------------------ ≥ 1.0
1 – ΣP u ⁄ φΣP c
and
Pc = π2EI/(Klu)2
In the previous expressions, Σ Pu and Σ Pc are the summa-
tions for all columns in a story. For frames not braced against
side-sway, both δb and δs should be computed. For frames
braced against side-sway, δs should be zero. In calculating P,
K should be computed according to Section 7.3.6.4(a) for δb
and according to Section 7.3.6.4(b) for δs.
In lieu of a more accurate calculation, EI in Eq. (7-41) may
be either
E c I g ⁄ 5 + E s I se
EI = -----------------------------------
- (7-42)
1 + βd
or conservatively
E c I g ⁄ 2.5
EI = ---------------------
- (7-43)
1 + βd
where βd is defined as the ratio of maximum factored dead
load moment to maximum factored total load moment, al-
ways positive (ACI 318).
In Eq. (7-39) Cm, for members braced against side-sway
and without transverse loads between supports, may be
Cm = 0.6 + 0.4M1/M2
but not less than 0.4. For all other cases, Cm should be 1.0
(ACI 318).
If the computations show that there is no moment at both
ends of a compression member, or that the computed end
eccentricities are less than (0.6 + 0.03h) in. [(15 + 0.03h)
mm], M2 in Eq. (7-38) should be based on a minimum eccen-
tricity of (0.6 + 0.03h) in. [(15 + 0.03h) mm] about each prin-
cipal axis separately. The ratio M1/M2 in Eq. (7-41) should
be determined as follows (ACI 318):
a. When the computed end eccentricities are less than (0.6
+ 0.03h) in. [(15 + 0.03h) mm], computed end
moments may be used to evaluate M1/M2.
b. If computations show that there is essentially no
moment at both ends of a compression member, the
ratio M1/M2 should be equal to one.
343R-87
(7-38) When compression members are subject to bending about
principal axes, the moment about each axis should be magni-
fied by δb and δs and computed from corresponding condi-
tions of restraint about that axis. In structures which are not
braced against side-sway, the flexural members should be
(7-39) designed for the total magnified end moments of the
compression members at the joint.
When a group of compression members on one level
(7-40)
same superstructure and collectively resist the side-sway of
the structure, δb and δs should be computed for the member
group, as described in Section 7.3.6.5.
7.3.7 Shear strength required—The design of cross
sections subject to shear should be based on
(7-41)
Vu ≤ φNn (7-45)
where Vu is the factored shear force at the section under con-
sideration, and Vn is the nominal shear strength computed by
Vn = Vc + Vs (7-46)
where Vc is the nominal shear strength provided by the con-
crete in accordance with Section 7.3.8 or 7.3.9, and Vs is the
nominal shear strength provided by the shear reinforcement
in accordance with Section 7.3.11.
In determining the shear strength Vc whenever applicable,
the effects of axial forces due to creep, shrinkage, and
temperature changes, should be considered in restrained
members, and the effects of inclined flexural compression in
variable-depth members may be included.
The maximum factored shear force Vu at supports may be
computed in accordance with (c) or (d) below when both of
the following conditions (a) and (b) are satisfied (ACI 318):
a. The support reaction in the direction of the applied
shear introduces compression into the end regions of
the member.
b. No concentrated load occurs between the face of the
support and the location of critical section as defined
below in (c) or (d).
(7-44) c. For nonprestressed members, sections located less than
a distance d from the face of the support may be
designed for the same shear Vu as that computed at a
distance d.
d. For prestressed members, sections located less than a
distance h/2 from the face of the support may be
designed for the same shear Vu as that computed at a
distance h/2.
For deep flexural members, brackets, corbels, slabs and
footings, the recommendations of Section 7.3.14 through
7.3.16 should be followed.
7.3.8 Shear strength provided by concrete for nonpre-
stressed members
7.3.8.1 Simplified strength calculations—Shear strength
Vc should be computed by the provisions of (a) through (d)
below, unless a more detailed calculation is made (ACI 318).
a. For members subject to shear and flexure only
343R-88
Vc = 2 f c ′ bwd (0.17 f c ′ bwd)
b. For members subject to axial compression
Vc = 2(1 + 0.005 Nu/Ag)
[Vc = 0.17 (1 + 0.073Nu/Ag) f c ′ bwd]
with Nu/Ag expressed in psi (MPa).
c. For members subject to significant axial tension, the
shear reinforcement should be designed to carry the
total shear.
d. At sections of members where the torsional moment Tu
exceeds
2 2
φ ( 0.5 f c ′x y ) [ φ ( 0.04 f c ′x y ) ]
2 f c ′b w d
V c = -------------------------------------------------
1 + ( 2.5C t T u ⁄ V u )
0.17 f c ′b w d
V c = -------------------------------------------------
1 + ( 2.5C t T u ⁄ V u )
where x is the smaller overall dimension, and y is the larger
overall dimension of the rectangular cross section.
7.3.8.2 Detailed strength calculations—Alternatively,
shear strength Vc may be computed by the more detailed
provisions of (e) through (g) below (ACI 318):
e. For members subject to shear and flexure only
Vc = [1.9 f c ′ + 2500ρw(Vud/Mu)]bwd
{Vc = [0.16 f c ′ + 17.2ρw(Vud/Mu)]bwd}
but not greater than 3.5 f c ′ bwd or (0.29 f c ′ bwd). The
quantity Vud/Mu should not be greater than 1.0 for computing
Vc by Eq. (7-50), where Mu is the factored moment occurring
simultaneously with Vu at the section under consideration.
f. For members subject to axial compression, Eq. (7-50)
may be used to compute Vc with a modified moment
Mm, Mm substituted for Mu, and Vud/ Mu not limited to
1.0, where
Mm = Mu – Nu(4h – d)/8
However, Vc should not be greater than
V c = 3.5 f c ′ b w d 1 + N u ⁄ 500A g
( V c = 0.29 f c ′ b w d 1 + 0.29N u ⁄ A g )
MANUAL OF CONCRETE PRACTICE
(7-47) Quantity Nu/Ag should be expressed in psi (MPa). When
Mm as computed by Eq. (7-51), is negative, Vc should be
computed by Eq. (7-52).
g. For members subject to significant axial tension
f c ′ bwd
Vc = 2(1 + Nu/500Ag) f c ′ bwd
(7-48)
[Vc = 0.17 (1 + 0.29NuAg) f c ′ bwd] (7-53)
where Nu is negative for tension. Quantity Nu/Ag should be
expressed in psi (MPa).
7.3.9 Shear strength provided by concrete for prestressed
members
7.3.9.1 Basic strength calculation—For members with
an effective prestress force no less than 40 percent of the
tensile strength of the flexural reinforcement, the shear
strength Vc should be computed by equation (7-54), unless a
more detailed calculation is made (ACI 318)
-
2
Vc = (0.6 f c ′ + 700 Vud/Mu)bwd
[Vc = (0.05 f c ′ + 4.8 Vud/Mu)bwd] (7-54)
- (7-49)
2
But Vc need not be less than 2 f c ′ bwd (0.17 f c ′ bwd), nor
should Vc be greater than 5 f c ′ bwd (0.42 f c ′ bwd), or the
value given in Section 7.3.9.3. The quantity Vud/Mu should
not be greater than 1.0, where Mu is the factored moment
occurring simultaneously with Vu at the section considered.
When applying Eq. (7-54), d in the quantity Vud/Mu should
be the distance from the extreme compression fiber to the
centroid of the prestressed reinforcement.
7.3.9.2 Detailed strength calculations—Alternatively,
the shear strength Vc may be computed in accordance with
(a) through (c) below, where Vc should be the lesser of Vci or
(7-50) Vcw.7-3
a. Shear strength Vci should be computed by
Vci = 0.6 f c ′ bwd + Vd + ViMcr/Mmax
(Vci = 0.05 f c ′ bwd + Vd + ViMcr/Mmax) (7-55)
but Vci need not be less than
(7-51) 1.7 f c ′ bwd (0.14 f c ′ bwd)
where
Mcr = (I/yt) (6 f c ′ + fpe – fd)
(7-52) [Mcr = (I/yt) (0.5 f c ′ + fpe – fd)] (7-56)
 BRIDGE ANALYSIS AND DESIGN
and values of Mmax and Vi should be computed from the load
combination, causing the maximum moment to occur at the
section.
b. Shear strength Vcw should be computed by
Vcw = (3.5 f c ′ + 0.3fpc)bw d + Vp
[Vcw = (0.29 f c ′ + 0.3fpc)bw d + Vp] (7-57)
Vcw may also be computed as the shear force corre-
sponding to the dead load, plus live load that results in a prin-
cipal tensile stress of 4 f c ′ (0.33 f c ′ ) at the centroidal axis
of the member or at the intersection of flange and web, when
the centroidal axis is in the flange. In composite members,
the principal tensile stress should be computed using the
cross section that resists the live load.
c. In Eq. (7-55) and (7-57), d should be the distance from
the extreme compression fiber to the centroid of the
prestressed reinforcement or 0.8h, whichever is greater.
7.3.9.3 Strength reduction due to transfer length and
bonding—In pretensioned members, in which the section at a
distance h/2 from the face of the support is closer to the end of
the member than the transfer length of the prestressing
tendons, the reduced prestress should be considered when
computing Vcw. This value of Vcw should also be the maximum
limit for Eq. (7-54). Prestress force may be assumed to vary
linearly from zero at the end of a tendon to a maximum at a
distance from the end of the tendon equal to the transfer
length. The transfer length may be assumed to be 50 diameters
for strand and 100 diameters for single wire (ACI 318).
In pretensioned members, where the bonding of some
tendons does not extend to the end of the member, a reduced
prestress should be considered when computing Vc. The
value of Vcw calculated using the reduced prestress should
also be taken as the maximum limit for Eq. (7-54). The
prestress force, due to tendons for which bonding does not
extend to the end of the member, may be assumed to vary
linearly from zero at the point at which bonding commences to
a maximum at a distance from this point equal to the transfer
length. The transfer length may be assumed to be 50 diameters
for strand and 100 diameters for single wire (ACI 318).
7.3.10 Lightweight concrete shear strength—All provi-
sions for shear strength Vc and torsional moment strength Tc
are recommended for normal weight concrete. When light-
weight aggregate concrete is used, one of the modifications
below is recommended (ACI 318):
a. When fct is specified, provisions for Vc and Tc should be
modified by substituting fct /6.7 for f c ′ (0.08 f c ′ ), but
the value of fct /6.7 should not exceed fc ′
(0.08 f c ′ ).
b. When fct is not specified, all values of f c ′ (0.08 f c ′ )
affecting Vc , Tc , and Mcr should be multiplied by 0.75
for “all lightweight” concrete and by 0.85 for “sand-
lightweight” concrete. Linear interpolation may be
used when partial sand replacement is used.
7.3.11 Shear strength provided by shear reinforcement
343R-89
7.3.11.1 Types of shear reinforcement—Shear reinforce-
ment may consist of stirrups or closed ties perpendicular to the
axis of the member. In addition, for nonprestressed members,
shear reinforcement may also consist of:
a. Stirrups or closed ties making an angle of 45 deg or
more with the longitudinal reinforcement.
b. Longitudinal reinforcement with bent portion making
an angle of 30 deg or more with the longitudinal
tension reinforcement.
c. Combinations of stirrups or closed ties and bent longi-
tudinal reinforcement.
d. Spirals.
Design yield strength of shear reinforcement should not
exceed 60,000 psi (410 MPa).
Stirrups, closed ties, and other bars or wires used as shear
reinforcement should extend to a distance d from the
extreme compression fiber, and should be anchored at both
ends in accordance with Section 13.2, to develop the design
yield strength of the reinforcement.
7.3.11.2 Spacing limits for shear reinforcement—The
spacing of shear reinforcement placed perpendicular to the
axis of the member should not exceed d/2 in nonprestressed
members, 0.75h in prestressed members, or 24 in. (0.60 m).
Inclined stirrups and bent longitudinal reinforcement
should be spaced so that every 45 deg line, extending toward
the reaction from the mid-depth of the member d/2 to the
longitudinal tension reinforcement, is crossed by at least one
line of shear reinforcement.
When Vs exceeds 4 f c ′ bwd (0.33 f c ′ bwd), the maximum
spacings previously given should be reduced by one-half
(ACI 318).
7.3.11.3 Minimum shear reinforcement—A minimum
area of shear reinforcement should be provided in all rein-
forced concrete flexural members (both prestressed and
nonprestressed), where the factored shear force Vu exceeds
one-half the shear strength provided by the concrete φVc
except for:
a. Slabs and footings.
b. Beams with total depth not greater than 10 in. (0.25 m),
two-and-one-half times the thickness of the flange, or
one-half the width of the web, whichever is greater.
The previous minimum shear reinforcement recommenda-
tions may be waived if tests show that the required nominal
flexural and shear strengths can be developed when shear
reinforcement is omitted. Such tests should simulate the
effects of differential settlement, creep, shrinkage, and
temperature change, based on a realistic assessment of such
effects occurring in service (ACI 318).
Where shear reinforcement is recommended as above, or
by analysis, and where the factored torsional moment Tu
does not exceed
φ (0.5 f c ′ Σx2y) [φ (0.04 f c ′ Σx2y)]
the minimum area of shear reinforcement for prestressed
(except as provided for below) and nonprestressed members
should be computed as follows (ACI 318)
343R-90
Av = 50 bws/fy
(Av = 0.35 bws/fy)
where bw and s are in in. (mm).
For prestressed members with an effective prestress force
not less than 40 percent of the tensile strength of the flexural
reinforcement, the minimum area of shear reinforcement may
be computed by the lower of Eq. (7-58) or (7-59) (ACI 318)
A ps f pu s d
A v = -------
- ------ --- ------
80 f y d b w
Where the factored torsional moment Tu exceeds
φ(0.5 f c ′ Σx2y) or [φ (0.04 f c ′ Σx2y)], and where web rein-
forcement is recommended by this section or by analysis, the
minimum area of closed stirrups should be computed as
follows (ACI 318)
AvV + 2At = 50bws/fy
(Av + 2At = 0.35bws/fy)
7.3.11.4 Design of shear reinforcement—Where the
factored shear force Vu exceeds the shear strength φV, shear
reinforcement should be provided to satisfy Eq. (7-45) and
Eq. (7-46). The shear strength Vs should be computed in
accordance with the following paragraphs (ACI 318)
Vs = Avfyd/s
where Av is the area of shear reinforcement within a distance s.
When inclined stirrups are used as shear reinforcement
A v f y ( sin α + cos α )d
V s = --------------------------------------------------
- (7-62)
s
When shear reinforcement consists of a single bar or a
single group of parallel bars all bent up at the same distance
from the support
Vs = Avfysinα
but not greater than 3 f c ′ bwd (0.25 f c ′ bwd).
When shear reinforcement consists of a series of parallel
bent-up bars or groups of parallel bent-up bars at different
distance from the support, shear strength Vs should be
computed by Eq. (7-62). Only the center three-fourths of the
inclined portion of any longitudinal bent bar should be
considered effective for shear reinforcement.
Where more than one type of shear reinforcement is used
to reinforce the same portion of a member, the shear strength
Vs should be computed as the sum of the Vs values computed
for the various types.
Shear strength Vs should not be greater than 8 f c ′ bwd
(0.67 f c ′ bwd).
MANUAL OF CONCRETE PRACTICE
7.3.12 Combined shear and torsion strength for nonpre-
stressed members with rectangular, flanged, or box sections
(7-58) 7.3.12.1 General—Torsion effects should be included
with shear and flexure, where factored torsional moment Tu
exceeds φ(0.5 f c ′ Σx2y) [φ(0.04 f c ′ Σx2y)]. Otherwise,
torsion effects may be neglected (ACI 318).
For members with rectangular or flanged sections, the sum
Σx2y should be taken for the component rectangles of the
section, but the overhanging flange width used in the design
should not exceed three times the flange thickness (ACI 318).
A rectangular box section may be taken as a solid section,
(7-59)
if the wall thickness tw is at least x/4. A box section with a
wall thickness less than x/4, but greater than x/10, may also
be taken as a solid section, except that Σ x2y should be multi-
plied by 4tw /x. When h is less than x/10, the stiffness of the
wall should be considered. Fillets should be provided at inte-
rior corners of all box sections (ACI 318).
For a member having thin-walled open sections, consider-
ation of the torsion caused by restrained warping may be
necessary (ACI 318).
If the factored torsional moment Tu in a member is
(7-60)
required to maintain equilibrium, the member should be
designed to carry that torsional moment in accordance with
Section 7.3.12.2 (ACI 318).
In a statically indeterminate structure where reduction of
torsional moment in a member can occur due to a redistribu-
tion of internal forces, the maximum factored torsional
moment Tu may be reduced to φ(4 f c ′ Σx2y/3)
(7-61) [φ(0.33 f c ′ Σx2y/3)]. In such a case, the correspondingly
adjusted moments and shears in adjoining members should
be used in design (ACI 318).
In lieu of a more exact analysis, torsional loading from a slab
should be uniformly distributed along the member (ACI 318).
Sections located less than a distance d or b, whichever is
less, from the face of the support may be designed for the
same torsional moment Tu as that computed at that distance
d or b.
7.3.12.2 Torsional moment strength required—Design
of cross sections subject to torsion should be based on
(7-63) Tu ≤ φTn (7-64)
where Tu is the factored torsional moment at the section con-
sidered, and Tn is the nominal torsional moment strength
computed by
Tn = T c + T s (7-65)
where Tc is the nominal torsional moment strength provided
by the concrete in accordance with Section 7.3.12.3, and Ts
is the nominal torsional moment strength provided by torsion
reinforcement in accordance with Section 7.3. (ACI 318).
7.3.12.3 Torsional moment strength provided by
concrete—Torsional moment strength should be computed by
 BRIDGE ANALYSIS AND DESIGN
2 dance with Eq. (7-68) or (7-69). A minimum area of closed
0.8 f c ′Σx y
T c = -----------------------------------
0.4V 2
1 + ⎛ -------------u-⎞
⎝CT ⎠
t u
2
0.07 f c ′Σx y
T c = -----------------------------------
0.4V 2
1 + ⎛ -------------u-⎞
⎝CT ⎠
t u
For members subject to significant axial tension, torsion
reinforcement should be designed to carry the total torsional
moment, unless a more detailed calculation is made, in
which Tc is given by Eq. (7-66) and Vc given by Eq. (7-49) is
multiplied by
(1 + Nu/500Ag)[(1 + 0.29Nu/Ag)],
where Nu is negative for tension (ACI 318).
7.3.12.4 Torsion reinforcement recommendations—
Torsion reinforcement, where required, should be provided
in addition to reinforcement required to resist shear, flexure,
and axial forces.
Reinforcement required for torsion may be combined with
that required for other forces, provided that the area furnished is
the sum of individually required areas and the most restrictive
recommendations for spacing and placement are met.
Torsion reinforcement should consist of closed stirrups,
closed ties, or spirals, combined with longitudinal bars (ACI
318). The design yield strength of torsion reinforcement
should not exceed 60,000 psi (410 MPa).
Stirrups, closed ties, and other bars used as torsion rein-
forcement should extend to a distance d from the extreme
compression fiber and should be anchored according to
Section 13.2 to develop the design yield strength of the rein-
forcement.
Torsion reinforcement should be provided at least a
distance (d + b) beyond the point theoretically required.
The spacing of closed stirrups should not exceed the
smaller of (x1 + y1)/4, or 12 in. (0.3 m) (ACI 318).
The spacing of longitudinal bars not less than #3 (10 mm),
distributed around the perimeter of closed stirrups should not
exceed 12 in. (0.3 m). At least one longitudinal bar should be
placed in each corner of the closed stirrups (ACI 318).
7.3.12.5 Design of torsion reinforcement—Where the
factored torsional moment Tu exceeds the torsional moment
strength φTc torsion reinforcement should be provided to
satisfy Eq. (7-64) and (7-65), where the torsional moment
strength Ts should be computed as follows (ACI 318)
Ts = (At αt x1 y1 fy)/s
where At is the area of one leg of a closed stirrup resisting
torsion within a distance s, and αt = 0.66 + 0.33 (y1/x1), but
not more than 1.50. Longitudinal bars distributed around the
perimeter of the closed stirrups should be provided in accor-
343R-91
stirrups should be provided in accordance with Section
7.3.11.3.
The required area of longitudinal bars Al distributed
around the perimeter of the closed stirrups At should be
computed as follows (ACI 318)
Al = 2At (x1 + y1)/s (7-68)
(7-66)
or by
A l = 400xs
Tu
-------------- ⎛ -----------------------------------
-⎞ – 2A t ( x 1 + y 1 ) ⁄ s
f y T u + V u ⁄ ( 3C t )⎠
⎝
⎧ 2.76xs ⎛ Tu ⎞ ⎫
⎨ A l = ---------------- ⎝ ------------------------------------⎠ – 2A t ( x 1 + y 1 ) ⁄ s ⎬
⎩ f y T u + V u ⁄ ( 3C t ) ⎭
(7-69)
whichever is greater. The value of Al computed by Eq. (7-69)
need not exceed that obtained by substituting:
50bws/fy (0.35bws/fy) for 2At.
The torsional moment strength Ts should not exceed 4Tc.
7.3.13 Combined shear and torsion strength for
prestressed members—No codes or specifications presently
address torsion strength design for prestressed concrete
members. However, current proposed design procedures,
based on model analogies or test results, can provide a
rational approach applicable to bridge structures.
Design recommendations based on compression field
theory, including a bridge girder design example, are given
in Reference 7-6.
Skew bending theory is presented in Reference 7-7, along
with a generalized design procedure, and proposed modifica-
tions to existing codes to incorporate prestressed concrete.
7.3.14 Shear-friction
7.3.14.1 General—The provisions of this section should
be applied where it is appropriate to consider shear transfer
across a given plane, such as (ACI 318):
a. An existing or potential crack.
b. An interface between dissimilar materials.
c. An interface between two concretes cast at different
times.
The design of cross sections subject to shear transfer, as
described in this section, should be based on Eq. (7-45),
where Vn is calculated in accordance with the provisions of
Section 7.3.14.2.
A crack should be assumed to occur along the shear plane
considered. The required area of shear friction reinforcement
(7-67) Avf across the shear plane may be designed using either
Section 7.3.14.2 or any other shear transfer design methods
that result in a prediction of strength in substantial agreement
with the results of comprehensive tests. The provisions of
Section 7.3.14.2 should be followed for all calculations of
shear transfer strength.
343R-92 MANUAL OF CONCRETE PRACTICE
7.3.14.2 Shear-friction design method—When shear-
friction reinforcement is perpendicular to the shear plane, the
shear strength Vn should be computed as follows (ACI 318)
Vn = Avf fy μ (7-70)
where μ is the coefficient of friction listed below.
When shear-friction reinforcement is inclined to the shear
plane, and shear force produces tension in the shear-friction
reinforcement, the shear strength should be computed as
follows (ACI 318)
Vn = Avf fy (μ sinαf + cos αf) (7-71)
where αf is the angle between the shear-friction reinforcement
and the shear plane, which should never be less than 30 deg.
For both case, φVn should be equal to or greater than Vu.
The coefficient of friction μ in Eq. (7-70) and Eq. (7-71)
should be as follows (ACI 318):
Concrete placed monolithically 1.4λ
Concrete placed against hardened concrete with
the surface intentionally roughened as specified
below 1.0λ
Concrete placed against hardened concrete not
intentionally roughened 0.6λ
Concrete anchored to as-rolled structural steel by
headed studs or by reinforcing bars (see below) 0.7λ
where λ = 1.0 for normal weight concrete, 0.85 for “sand-
lightweight” concrete, and 0.75 for “all-lightweight”
concrete. Linear interpolation may be applied when partial
sand replacement is used.
Shear strength Vn should not be greater than 0.2fc′Ac
(0.107fc′Ac) or 800Ac (5.52Ac MPa) in kb (N), where Ac is
the area of the concrete section resisting shear transfer. The
design yield strength of shear-friction reinforcement should
not exceed 60,000 psi (410 MPa).
The net tension across the shear plane should be resisted
by additional reinforcement. Permanent net compression
across the shear plane may be taken as additive to the force
in the shear-friction reinforcement Avf fy when calculating the
required Avf.
Shear friction reinforcement should be appropriately
placed along the shear plane and should be anchored to
develop the specified yield strength on both sides by embed-
ment, hooks, or welding to special devices.
For the purpose of shear-friction, when concrete is placed
against previously placed concrete, the interface for shear
transfer should be clean and free of laitance. If μ is assumed
equal to 1.0λ, the interface should be roughened to a full
amplitude of approximately 1/4 in. (6 mm).
When shear is transferred between as-rolled steel and
concrete using headed or welded reinforcing bars, the steel
should be clean and free of paint.
7.3.15 Horizontal shear design for composite concrete
flexural members—Provision should be made for full
transfer of horizontal shear forces at contact surfaces of
interconnected elements.
7.3.15.1 Calculations for shear—The design horizontal
shear stress at any cross section may be computed by
vdh = V/(bvd) (7-72)
where d is the depth of the entire composite section.
However, horizontal shear also may be investigated by
computing, in any segment not exceeding one-tenth of the
span, the actual change in compressive or tensile force to be
transferred, and provisions made to transfer that force as
horizontal shear between interconnected elements.
7.3.15.2 Allowable shear—Horizontal shear may be
transferred at contact surfaces using the permissible hori-
zontal shear stress vh stated below:
a. When minimum ties are provided and contact surfaces
are clean and free of laitance, but not intentionally
roughened, shear stress vh should not exceed 36 psi.
b. When minimum ties are provided and contact surfaces
are clean, free of laitance, and intentionally roughened
to a full amplitude of approximately 1/4 in., shear stress
vh should not exceed 160 psi.
c. For each percent of tie reinforcement crossing the
contact surface in excess of the minimum specified
below, permissible vh may be increased by 72fy /40,000
psi (0.492fy /276 MPa).
7.3.15.3 Ties for horizontal shear—A minimum area of
tie reinforcement should be provided between intercon-
nected elements. Tie area should not be less than 50bv s/fy ,
and tie spacing should not exceed four times the least web
width of support element, or 24 in. (0.6 m).
Ties for horizontal shear may consist of single bars or
wire, multiple leg stirrups, or vertical legs of welded wire
fabric (smooth or deformed). All ties should be adequately
anchored into interconnected elements by embedment or
hooks.
7.3.16 Special shear provisions for deep flexural
members—The provisions of this section are recommended
for members with ln/d of less than 5 and loaded at the top or
compression face.
The design of deep flexural members for shear should be
based on Eq. (7-45) and (7-46), where shear strength Vc
should be calculated by Eq. (7-73) or (7-77), and shear
strength Vs should be calculated by Eq. (7-75) (ACI 318).
Shear strength Vn for deep flexural members should not be
greater than 8 f c ′ bwd (0.67 f c ′ bwd), when ln/d is less than
2. When ln /d is between 2 and 5
Vn = 0.667(10 + ln/d) f c ′ bwd
[Vn = 0.055(10 + ln/d) f c ′ bwd] (7-73)
The critical section for shear, measured from the face of
the support, should be at a distance of 0.15ln for uniformly,
loaded beams and 0.5a for beams with concentrated loads,
but not greater than d.
 BRIDGE ANALYSIS AND DESIGN
Unless a more detailed calculation is made using Eq. (7-75),
Vc should be computed by
Vc = 2 f c ′ bwd
(Vc = 0.17 f c ′ bwd)
Shear strength Vc may be calculated by
Vc = (3.5 - 2.5Mu/Vud)(1.9 f c ′ + 2500ρwVud/Mu)bwd
[Vc = (3.5 - 2.5Mu/Vud)(0.16 f c ′ +
17ρwVud/Mu)bwd]
except that the term (3.5 – 2.5Mu/Vud) should not exceed 2.5,
and V should not be greater than 6 f c ′ bwd or (0.5 f c ′ bwd).
Mu is the factored moment occurring simultaneously with Vu
at the critical section as previously defined.
Where the factored shear force Vu exceeds the shear
strength φVc, shear reinforcement should be provided to
satisfy Eq. (7-45) and (7-46), where the shear strength Vs
should be computed by (ACI 318)
A 1 + l n ⁄ d⎞ A vh ⎛ 11 – l n ⁄ d⎞
V s = -----v ⎛ -------------------
- + -------- ----------------------- f y d
s ⎝ 12 ⎠ s 2 ⎝ 12 ⎠
where Av is the area of shear reinforcement perpendicular to
the flexural tension reinforcement within a distance s, and
Avh is the area of shear reinforcement parallel to the flexural
reinforcement within a distance s2.
The area of shear reinforcement Av should not be less than
0.0015bs2, and s should not exceed d/5, or 18 in. (0.5 m). The
area of shear reinforcement Avh should not be less than
0.0025bs2, and s2 should not exceed d/3, or 18 in. (0.5 m).
The shear reinforcement required at the critical section, as
defined above, should be used throughout the span.
7.3.17 Special shear provisions for brackets and
corbels—The provisions of this section are recommended
for brackets and corbels with a shear span-to-depth ratio la/d,
not greater than 1.0, and subject to a horizontal tensile force
Nuc not greater than Vu. The distance la should be measured
at the face of the support. The depth at the outside edge of the
bearing area should not be less than 0.5d.
The section at the face of the support should be designed to
simultaneously resist a shear Vu, a moment [Vula + Nuc(h – d)],
and a horizontal tensile force Nuc.
For all design calculations in this section, the strength
reduction factor φ should be equal to 0.85.
The design of shear-friction reinforcement Avf to resist
shear Vu should be in accordance with Section 7.3.14. Shear
strength Vn should be limited according to (a) and (b) below
(ACI 318):
a. For normal weight concrete, Vn should not be greater than
0.2fc′bwd (0.017fc bwd) or 800bwd (5.52bwd), in.-lb (N).
343R-93
b. For “all-lightweight” or “sand-lightweight” concrete,
Vn should not be greater than
(0.2 – 0.07la/d)fc′bwd [(5.52 – 1.93la/d)fc′bwd),
or (800 - 280la/d)bwd in.-lb (N).
(7-74)
The reinforcement Af to resist the moment [Vula – Nuc (h – d)],
should be computed in accordance with Sections 7.3.3
and 7.3.4.
Reinforcement An, to resist tensile force Nuc, should be
determined from Nuc ≤ φAn fy. The tensile force Nuc should
not be less than 0.2Vu, unless special provisions are made to
avoid tensile forces. The tensile force Nuc should be regarded
as a live load, even when the tension results from creep,
(7-75) shrinkage, or temperature change. The area of primary
tension reinforcement As should be equal to the greater of (Af
+ An) or (2Avf /3 + An) (ACI 318).
Closed stirrups or ties parallel to As with a total area Ah not
less than 0.5 (As – An) and spaced no further apart than the
smaller of d/4 or 12 in. (305 mm) center to center, should be
uniformly distributed within the three-quarters of the effective
depth adjacent to As. The anchorage requirements of Section
13.2.13 should be followed for this reinforcement also.
The ratio ρ = As/bd should not be less than 0.04 (fc′/fy).
At the front face of the bracket or corbel, the primary
tension reinforcement As should be anchored by one of the
(7-76)
following methods (ACI 318)
a. By a structural weld to a transverse bar of at least equal
size, with the weld designed to develop the specified
yield strength fy of the As bars.
b. By bending the primary tension bars As back to form a
horizontal loop.
c. By some other means of positive anchorage.
The bearing area of the load on the bracket or corbel
should not project beyond the straight portion of the primary
tension bars As or beyond the interior face of the transverse
anchor bar, if one is provided.
7.3.18 Special shear recommendations for slabs and footings
7.3.18.1 General—The shear strength of slabs and foot-
ings in the vicinity of concentrated loads or reactions should
be governed by the more severe of the two conditions
described in (a) and (b) below (ACI 318):
a. Beam action for slabs or footings, with a critical section
extending in a plane across the entire width, and
located at a distance d from the face of the concentrated
load or the reaction area. For this condition the slab or
footing should be designed in accordance with Sections
7.3.7 through 7.3.11.
b.Two-way action for slabs or footings, with a critical
section perpendicular to the plane of the slab, and
located so that its perimeter bo is a minimum. However,
the critical section need not approach closer than d/2 to
the perimeter of the concentrated load or reaction area.
For this condition, the slab or footing should be
designed in accordance with Sections 7.3.18.2 and
7.3.18.3.
343R-94 MANUAL OF CONCRETE PRACTICE
7.3.18.2 Slabs and footings without shear reinforce-
ment—The design of slabs or footings for two-way action
should be based on Eq. (7-42), where shear strength Vn should
not be greater than the shear strength Vc computed by Eq.
(7-77), unless shear reinforcement is provided in accordance
with Section 7.3. For nonprestressed slabs and footings
Vc = (2 + 4/βc) f c ′ bod
[Vc = 0.08 (2 + 4/βc) f c ′ bod] (7-77)
but not greater than 4 f c ′ bod (0.33 f c ′ bod). βc is the ratio
of long side to short side of the concentrated load or reaction
area, and bo is the perimeter of the critical section as defined
above.
7.3.18.3 Slabs and footings with shear reinforcement—It
is generally not economically feasible to choose a slab or
footing thin enough to require shear reinforcement. However,
shear reinforcement consisting of bars may be used in slabs
and footings in accordance with (a) through (e) below:
a. Shear strength Vn should be computed by Eq. (7-46),
where the shear strength Vc should be in accordance
with Subsection (d) below, and shear strength Vs should
be in accordance with (e) below.
b. Shear strength Vn should not be greater than 6 f c ′ bod
(0.5 f c ′ bod), where bo is the perimeter of the critical
section as defined in (c) below.
c. Shear strength should be investigated at the critical
section as defined in Section 7.3.18.1 and at successive
sections more distant from the support.
d. Shear strength Vc at any section should not be greater
than 2 f c ′ bod (0.17 f c ′ bod), where bo is the perimeter
of the critical section as defined in (c) above.
e. Where the factored shear force Vu exceeds the shear
strength φVc as given in (d) above, the recommended
area Av and the shear strength Vs of the shear reinforce-
ment should be calculated in accordance with Section
7.3.11 and anchored in accordance with Section 13.2.
7.3.19 Transfer of moment to columns—When gravity
load, wind, earthquake, or other lateral forces cause transfer
of moment at connections of framing elements to columns,
the shear resulting from moment transfer should be consid-
ered in the design of the lateral column reinforcement.
Lateral reinforcement not less than that computed by
Eq. (7-55) should be provided within the connections of the
framing elements to the columns, except for connections not
part of a primary seismic load-resisting system that are
restrained on four sides by beams or slabs of approximately
equal depth.
7.3.20 Bearing strength—The design bearing strength of
concrete should not exceed φ(0.85fc′A1), except as follows
(ACI 318):
a. When the supporting surface is wider on all sides than
the loaded area, the design bearing strength on the
loaded area may be multiplied by A2/A1, but not more
than 2.
b. When the supporting surface is sloped or stepped, A2
may be taken as the area of the lower base of the largest
frustrum of a right pyramid or cone, contained wholly
within the support, having for its upper base the loaded
area, and having side slopes of 1 vertical to 2 hori-
zontal.
This section does not apply to post-tensioning anchorages.
RECOMMENDED REFERENCES
The documents of the various standards-producing organi-
zations referred to in this report are listed below with their
serial designation, including year of adoption or revision.
The documents listed were the latest effort at the time this
report was written. Since some of these documents are
revised frequently, generally in minor detail only, the user of
this report should check directly with the sponsoring group
if it is desired to refer to the latest revision.
American Association of State Highway and Transportation
Officials
Standard Specifications for Highway Bridges, Thirteenth
Edition, 1983
American Concrete Institute
Recommendations of Joint Task Group of ACI Committees
343 and 348 on Load and Resistance Factors for Concrete
Bridges
318-83 Building Code Requirements for Reinforced
Concrete
318R-83 Commentary on Building Code Requirements for
Reinforced Concrete
American Railway Engineering Association
Manual for Railway Engineering
Prestressed Concrete Institute
MNL-116-85 Manual for Quality Control for Plants and
Production of Precast and Prestressed
Concrete Products, Third Edition
CITED REFERENCES
7-1. Naaman, A. E., “Partially Prestressed Concrete:
Review and Recommendations,” PCI Journal, V. 30, No. 6,
Nov.-Dec. 1985, pp. 30-71.
7-2. “Bridge Design Practice—Load Factor,” California
Department of Transportation, Sacramento, CA, 1983, pp. 2-
171 to 2-173.
7-3. Naaman, A. E., and Harajli, M. H., “Evaluation of the
Ultimate Steel Stress in Partially Prestressed Flexural
Members,” PCI Journal, V. 30, No. 5, Sept.-Oct. 1985, pp.
54-81.
7-4. Portland Cement Association, “Biaxial and Uniaxial
Capacity of Rectangular Columns,” EB031.01D, Skokie, IL,
1967.
7-5. Portland Cement Association, “Notes on ACI 318-
83,” EB070.04D, Skokie, IL, 1984.
 BRIDGE ANALYSIS AND DESIGN 343R-95

7-6. Collins, M. P., and Mitchell, D., “Shear and Torsion 7-7. Hsu, T. T. C., Torsion of Reinforced Concrete, Van
Design of Prestressed and Nonprestressed Concrete Beams,” Nostrand Reinhold Co., New York, 1984.
PCI Journal, V. 25, No. 5, Sept.-Oct. 1980, pp. 32-100.
343R-96 MANUAL OF CONCRETE PRACTICE

Completed box girder span at Corpus Christi, Texas (photo courtesy of Texas Highway Department).

CHAPTER 8—SERVICE LOAD ANALYSIS AND DESIGN

8.1—Basic assumptions e. In doubly reinforced flexural members, an effective

8.1.1 Nonprestressed members
For investigation of service load stresses, the straight line
theory of stress and strain in flexure should be used and the
following assumptions should be made:
a. A section plane before bending remains plane after
bending. Stress-strain relationship of concrete is a
straight line under service loads within permissible
service stresses.
b. Strain in reinforcement and concrete should be assumed
directly proportional to the distance from the neutral
axis, except for deep flexural members with overall
depth to clear span ratios greater than two-fifths for
continuous spans and four-fifths for simple spans, a
nonlinear distribution of strain should be considered.
c. Tensile strength of concrete should be neglected in
flexural calculations of reinforced concrete, except
when meeting requirements of Section 8.7.
d. Modular ratio n = Es/Ec may be taken as the nearest
whole number, but not less than 6. Except in calcula-
tions for deflections, value of n for lightweight concrete
should be assumed to be the same as for normal weight
concrete of the same strength.
modular ratio of 2 Es/Ec should be used to transform
compression reinforcement for stress computations.
Compressive stress in such reinforcement should not
exceed the permissible tensile stress.
8.1.2 Prestressed members—For investigation of stresses
at transfer of prestress, at service loads, and at cracking
loads, straight line theory may be used with the following
assumptions:
a. Strain varies linearly with depth through the entire load
range.
b. Before cracking, stress is linearly proportional to strain.
c. At cracked sections, concrete resists no tension.
d. In calculations of section properties, the transformed
area of bonded reinforcement may be included in
pretensioned members and in post-tensioned members
after grouting.
8.2—Serviceability requirements
Moments, shears, and axial and torsional forces should be
obtained by the theory of elastic analysis or as recommended
in Chapter 10.
8.2.1 Nonprestressed flexural members—For members
 BRIDGE ANALYSIS AND DESIGN 343R-97

designed with reference to load factors and strengths by

Chapter 7, stresses at service load should be limited to satisfy
the recommendations for fatigue in Section 8.3 and for distri-
bution of reinforcement in Section 8.4. The recommenda-
tions for deflection control in Section 8.5 should also apply.
For nonprestressed members designed with reference to
service loads and allowable stresses, the serviceability
requirements recommended in Section 8.4 and the strength
requirements recommended in other chapters of this docu- (a) Lug profile
ment may be assumed satisfied, if the service load stresses
are limited to the values given in Sections 8.3 and 8.8.
8.2.2 Prestressed members—Design should be based on
strength design and on behavior at service conditions at all
load stages that may be critical during the life of the struc-
ture, from the time the prestress is first applied. Stresses at
service conditions should be limited to values given in
Section 8.7.

8.3—Fatigue of materials
8.3.1 Reinforcing bars—For straight hot-rolled bars with
no welds and with no stress raisers more severe than defor-
mations meeting the requirements of ASTM reinforcing bar
specifications, the range between a maximum tension stress
and minimum stress caused by live load plus impact, at
service load, should not exceed
ff = 21 – 0.33fmin + 8(r/h)
[ff = 145 – 0.33fmin + 55(r/h)] (8-1)
where
ff = stress range calculated as algebraic difference
between maximum stress and minimum stress
when tension is positive and compression is
negative, ksi (MPa).
fmin = algebraic minimum stress level, tension positive,
compression negative, ksi (MPa).
r/h = ratio of base radius to height of rolled-on trans-
verse deformation; when actual value is not
known, 0.3 is recommended (see Fig. 8.3).
Greater stress ranges may be used if determined by fatigue
tests on similar bars. These recommendations are based on
References 8-1 and 8-2.
For bars containing welds conforming to requirements of
AWS D1.4, the range between a maximum tension stress and
minimum stress caused by live load plus impact at service
load in the reinforcing element in the vicinity of the weld,
should not exceed that allowed above or 18 ksi (124 MPa).8-1
Reinforcing bars should not be welded without regard to
steel weldability and proper welding procedures.
Fatigue stress limits need not be considered for concrete
deck slabs with main reinforcement perpendicular to traffic
and designed in accordance with the approximate methods
given under Section 10.4.8-3—8-7
8.3.2 Prestressing steel—The stress range in prestressed
reinforcement that may be imposed on minimum stress
levels up to 60 percent of the tensile strength, should not
exceed the following:
(b) Lugs and ribs
Fig. 8.3—Bar geometry
Strand and bars 0.10fpu
Wires 0.12fpu
where fpu is the ultimate strength of the prestressing steel. For
additional information see References 8-2 and 8-8.
8.4—Distribution of reinforcement in flexural
members
8.4.1 General—Only deformed reinforcement should be
considered effective for principal reinforcement, except that
plain bars or smooth wire may be used as spirals or reinforce-
ment ties for confinement. To control flexural cracking of
the concrete, tension reinforcement should be well distrib-
uted in zones of maximum tension.
When the design yield strength fy for tension reinforce-
ment exceeds 40,000 psi (276 MPa), cross sections of
maximum tension stress should be proportioned so that the
calculated stress in the reinforcement at service load fs in
kips/in.2 (MPa) does not exceed the value computed by
z
f s = --------------------- (8-2)
1⁄3
( dc A )
but fs should not be greater than 0.6fy
where
A = Effective tension area in in.2 (mm2) of concrete
surrounding the main tension reinforcing bars and
having the same centroid as the reinforcement,
divided by the number of bars or wires. When the
343R-98
main reinforcement consists of several bar or wire
sizes, the number of bars or wires should be
computed as the total steel area divided by the area
of largest bar or wire used.
dc = Thickness of concrete cover measured from the
extreme tension fiber to the center of the bar
located closest thereto in in. (mm).
The quantity z in Eq. (8-2), should not exceed 175 kips/in.
(30.2 MN/m) for members in moderate exposure conditions
and 145 kips/in. (23.1 MN/m) for members in severe expo-
sure conditions. Where members are exposed to very aggres-
sive exposure or corrosive environments, such as deicer
chemicals, the denseness and nonporosity of the protecting
concrete should be increased, or other protection, such as a
waterproof protecting system, should be provided in addition
to satisfying Eq. (8-2).
8.4.2 T-beam flanges—Where flanges of T-beam
construction are in tension, part of the flexural tension rein-
forcement should be distributed over the effective flange
width, as defined in Section 7.2., or a width equal to one-
tenth the span, whichever is smaller. If the effective flange
width exceeds one-tenth the span, additional longitudinal
reinforcement with area not less than 0.4 percent of the
excess slab area, should be provided in the outer portions of
the flange.
8.4.3 Deep members—If depth d of a member exceeds 36 in.
(0.9 m), longitudinal crack control reinforcement should be
provided near the side faces of the member and distributed in
the zone of flexural tension. Area of reinforcement in in.2
(mm2) should be computed by
For depths d less than 100 in. (2.5 m)
As = 0.00024 (d – 30)Ae
[As = 0.00967 (d – 0.76)Ae]
For depths d greater than 100 in. (2.5 m)
As = (0.011 + 0.000058d)Ae
[As = (0.011 + 0.00228d)Ae]
Table 8—Recommended minimum thickness for
constant depth members*
Superstructure type Minimum depth†
ft
Bridge slabs with main L + 10-
reinforcement parallel or --------------
perpendicular to traffic 30
but not less than 0.542 (0.164)
T-girders L+9
------------
18
Box girders
L + 10
---------------
20
* When variable depth members are used, table values may be adjusted to account for
change in relative stiffness of positive and negative moment sections.
† Recommended values for continuous spans. Simple spans should have about a 10
percent greater depth.
L = Span length of member in ft (m).
MANUAL OF CONCRETE PRACTICE
where
Ae = Effective tension area of concrete along side face
of member surrounding the crack control rein-
forcement and having the same centroid as that
reinforcement with a depth equal to d/2.
Crack control reinforcement should be uniformly distrib-
uted along the side faces within the mid-depth d/2 on the
tension side of the member, with spacing not more than d/10
or 12 in. (0.3 m). Such reinforcement may be included in
strength computation if a strain compatibility analysis is
made to determine stresses in individual bars or wires.8-9
8.5—Control of deflections
8.5.1 General—Flexural members of bridge structures
should be designed to have adequate stiffness to limit deflec-
tions or any deformations which may adversely affect the
strength or serviceability of the structure at service load.
8.5.2 Superstructure depth limitations—The minimum
thicknesses stipulated in Table 8 are recommended,
unless computation of deflection indicates that lesser thick-
ness may be used without adverse effects.
8.5.3 Nonprestressed members
8.5.3.1 Computation of immediate deflection—Deflec-
tions which occur immediately on application of load should
be computed by the usual methods or formulas for elastic
deflections. Unless values are obtained by a more compre-
hensive analysis, immediate deflections should be computed
using the modulus of elasticity Ec for concrete as recom-
mended in Section 3.3 for normal or lightweight concrete
and using an effective moment of inertia Ie for simple spans
given by Eq. (8-3) or in accordance with the approximate
method given in Reference 8-10, but not greater than Ig.
Ie = (Mcr/Ma)3Ig + (1 – (Mcr/Ma)3)Icr (8-5)
(8-3)
where
Ie = Effective moment of inertia for computation of
deflection
Icr = Moment of inertia of cracked section with rein-
forcement transformed to concrete
(8-4)
Ig = Moment of inertia of gross section about the cent-
roidal axis neglecting the reinforcement
Ma = Maximum moment in member at stage for which
deflection is being computed
Mcr = Cracking moment = frIg/yt
yt = Distance from the centroidal axis of gross section,
m
neglecting the reinforcement, to the extreme fiber
L + 3-
----------- in tension
30
fr = Modulus of rupture of concrete from tests, or if
data is not available, the values given in Section
L + 2.75
-------------------- 8.7.1 may be used
18
For continuous spans, the effective moment of inertia may
L+3
------------ be taken as the average of the values obtained from Eq. (8-3)
20
for the critical positive and negative moment sections.
8.5.3.2 Computation of long-time deflections—In order to
obtain the total deflections, the additional long-time deflection
for both normal and lightweight concrete flexural members may
 BRIDGE ANALYSIS AND DESIGN 343R-99

be estimated by multiplying the immediate deflection, caused Tension in members without bonded auxiliary rein-
by the sustained load considered, by the factor forcement (nonprestressed or prestressed) in the tension
zone - 3 f ci ′ (0.25 f ci ′ ).
l = ξ/(1 + 50ρ′) (8-6) Where the calculated tensile stress exceeds this value,
bonded reinforcement should be provided to resist the total
where ρ′, the compression steel ratio, is the value at midspan tensile force in the concrete, computed on the assumption of
for simple and continuous spans and at support for cantile- an uncracked section. Stress in prestressing steel due to the
vers. Time-dependent factor ξ for sustained loads may be temporary jacking force should not exceed 0.80fpu, where fpu
equal to: is the ultimate strength of the prestressing steel, but not
5 years or more 2.0 greater than the maximum value recommended by the manu-
12 months 1.4 facturer of the steel or of the anchorage.
6 months 1.2 8.6.2 Service load stresses—Flexural stresses in concrete
3 months 1.0 at service load, after allowance for all prestress losses,
8.5.4 Prestressed members should not exceed the following:
8.5.4.1 Computation of immediate deflection—Deflec- Compression — 0.40fc′.
tions which occur immediately on application of load should Tension in precompressed tension zone:
be computed by the usual methods or formulas for elastic With bonded auxiliary reinforcement to control cracking
deflections, using the moment of inertia of the gross concrete — 6 f c ′ (0.5 f c ′ ).
section for uncracked sections. With bonded auxiliary reinforcement to control cracking,
8.5.4.2 Computation of long-time deflection—The addi- but exposed to corrosive environment or severe exposure
tional long-time deflection should be computed, taking into conditions — 3 f c ′ (0.25 f c ′ ).
account the stresses in the concrete and steel under the
Without bonded auxiliary reinforcement — 0.
sustained load, including the effects of creep and shrinkage
Stresses in pretensioning tendons immediately after
of the concrete and relaxation of the steel.
transfer or in post-tensioning tendons immediately after
anchoring should not exceed 0.70fpu.
8.6—Permissible stresses for prestressed flexural
members Post-tensioned anchorage stress should not exceed 3000
8.6.1 Temporary stresses—Flexural stresses in concrete psi (20.7 MPa).
immediately after transfer and before losses due to creep and The permissible stresses previously recommended may be
shrinkage, should not exceed the following stresses based on exceeded, if it is shown by tests or analysis that performance
fci′, the compressive strength of concrete at the time of initial will not be impaired.
prestress:
Compression: 8.7—Service load design
Pretensioned 0.60fci′ 8.7.1 Flexure—Nonprestressed members may be designed
Post-tensioned 0.55fci′ using service loads and allowable stresses. The stresses in

Table 8.7.1—Allowable stresses

Description Basic value
For normal weight concrete psi MPa
Flexure
Extreme fiber stress in compression fc 0.4 fc′ 0.4 fc′
Extreme fiber stress in tension (plain concrete)* ft 1.6 f c ′ 0.13 f c ′
Modulus of rupture* fr 7.5 f c ′ 0.62 f c ′
Shear
Beams:
Shear carried by concrete* vc 0.95 f c ′ 0.08 f c ′
Maximum shear carried by concrete plus shear reinforcement v 5 fc ′ 0.42 f c ′
Slabs and footings:
Shear carried by concrete* vc 1.8 f c ′ 0.15 f c ′
Maximum shear carried by concrete plus shear reinforcement v 3 fc ′ 0.25 f c ′
Bearing on loaded area fb 0.3fc′ 0.3fc′
Bearing on loaded area subjected to high edge stresses due to fb 0.225fc′ 0.225fc′
deflection or eccentric loading
* When lightweight aggregate concretes are used, the allowable stresses should be multiplied by 0.75 for “all-lightweight” concrete, and 0.85 for “sand-lightweight” concrete.
Linear interpolation may be used when partial sand replacement is used.
343R-100 MANUAL OF CONCRETE PRACTICE
concrete and reinforcement in flexure should not exceed the
following (see also Table 8.7.1):
Extreme fiber stress in compression fc ................... 0.40fc′
Extreme fiber stress in tension for
plain concrete ft ..................................................... 0.21fr
fr = modulus of rupture from tests. If data are not available,
fr may be assumed to be:
Normal weight concrete ..................... 7.5 f c ′ (.623 f c ′ )
“Sand-lightweight” concrete .............. 6.3 f c ′ (.523 f c ′ )
“All-lightweight” concrete ................. 5.5 f c ′ (.456 f c ′ )
For a detailed summary of allowable shear stress vc, see
Section 8.7.4.2.
The bearing stress fb on the loaded area should not exceed
0.30fc′, except when the supporting surface is wider on all
sides than the loaded area. The allowable bearing stress on
the loaded area may be increased by A2/A1, but not by more
than 2.
When the supporting surface is sloped or stepped, A2 may
be taken as the area of the lower base of the largest frustum
of the right pyramid or cone contained wholly within the
support and having for its upper base the loaded area, and
having side slopes of 1 vertical to 2 horizontal.
When the loaded area is subjected to high edge stresses
due to deflection or eccentric loading, the allowable bearing
stress on the loaded area, including any increase due to the
supporting surface being larger than the loaded area, should
be multiplied by a factor of 0.75.
8.7.2 Development of reinforcement—The calculated
tension or compression reinforcement at each section should
be developed on each side of that section by proper embed-
ment length, end anchorage, or for tension only, hooks.
Development of reinforcement should be computed by
procedures recommended in Chapter 13.
8.7.3 Compression members—The combined axial load
and moment capacity of compression members should be 35
percent of that computed in accordance with the recommen-
dations of Section 7.3.5.
Slenderness effects should be included according to the
recommendations of Section 7.3.6. The term Pu in Eq. (7-36)
should be replaced by 2.85 times the design axial load. In using
the provisions of Sections 7.3.5 and 7.3.6, φ should be 1.0.
8.7.4 Shear
8.7.4.1 General—The design shear stress should be
computed by
V Concrete
v = ---------
bw d
where bw is the width of web and d is the distance from the
extreme compression fiber to the centroid of the longitudinal
tension reinforcement. For a circular section, bw should be the
diameter and d need not be less than the distance from the
extreme compression fiber to the centroid of the longitudinal
reinforcement in the opposite half of the member. Whenever
applicable, effect of torsion should be included in accordance
with the recommendations in Reference 8-11 and ACI 318.
When the reaction in the direction of the applied shear
introduces compression into the end regions of member,
sections located less than a distance d from the face of the
support may be designed for the same shear v as that at a
distance d.
8.7.4.2 Concrete—The shear stress carried by the
concrete vc should not exceed 0.95 f c ′ (0.08 f c ′ ). When v
exceeds vc, shear reinforcement should be provided.
8.7.4.3 Reinforcement—Shear reinforcement should
conform to the general requirements of Section 7.3.
When shear reinforcement perpendicular to the axis of the
member is used, the required area should be computed by
( v – v c )b w s
A v = --------------------------
- (8-8)
fs
When (v – vc) exceeds 2fc′ (0.17fc′), the maximum spacing
given in Section 7.3.7 should be reduced by one-half. The
maximum shear carried by concrete, plus shear reinforce-
ment, should not exceed 5 f c ′ (0.42 f c ′ ).
8.7.4.4 Deep members—For deep members, brackets,
corbels, and walls, the special provisions of ACI 318,
Chapter 11 should be used.
8.8—Thermal effects
Stresses or movements due to seasonal and diurnal
temperature changes should be provided for in the design as
recommended in Section 5.4.
RECOMMENDED REFERENCES
The documents of the various standards-producing organi-
zations referred to in this report are listed below with their
serial designation, including year of adoption or revision.
The documents listed were the latest effort at the time this
report was written. Since some of these documents are
revised frequently, generally in minor detail only, the user of
this report should check directly with the sponsoring group
if it is desired to refer to the latest revision.
American Association of State Highway and Transportation
Officials
Standard Specifications for Highway Bridges, Thirteenth
Edition, 1983.
American Concrete Institute
318-83 Building Code Requirements for Reinforced
(8-7)
American Welding Society
D1.4-79 Structural Welding Code for Reinforcing Steel
CITED REFERENCES
8-1. Helgason, T.; Hanson, J. M.; Somes, N. J.; Corley, W.
G.; and Hognestad, E., “Fatigue Strength of High-Yield
Reinforcing Bars,” Report to the Highway Research Board,
National Cooperative Highway Research Program, Mar.
1975, 371 pp. Also NCHRP Digest No. 73, June 1975, 6 pp.
 BRIDGE ANALYSIS AND DESIGN
8-2. ACI Committee 215, “Considerations for Design of
Concrete Structures Subjected to Fatigue Loading,” ACI
JOURNAL, Proceedings V. 71, No. 3, Mar. 1974, pp. 97-121.
8-3. Hawkins, N. M., “Fatigue Design Considerations for
Concrete Bridge Decks,” ACI JOURNAL, Proceedings V. 73,
No. 2, Feb. 1976, pp. 104-115.
8-4. Rabbat, B. G., et al., “Fatigue Tests of Pretensioned
Girders with Blanketed and Draped Strands,” PCI Journal,
V. 24, No. 4, July-Aug. 1979, pp. 88-114 (also PCA
RD062.01D).
8-5. Helgason, T., et al., “Fatigue Strength of High-Yield
Reinforcing Bars,” NCHRP Report No. 164, Transportation
Research Board, Washington, D.C., 1976, 30 pp. (also PCA
RD 045.01E).
8-6. Corley, W. G., Hanson, J. M. and Helgason, T.,
“Design of Reinforced Concrete for Fatigue,” Journal of the
Structural Division, ASCE, June 1978, pp. 921-932 (also
343R-101
PCA RD059.01D).
8-7. “Fatigue of Steel and Concrete Structures,” IABSE V.
37, 1982, Lausanne, Switzerland, 895 pp.
8-8. Okada, K., et al., “Fatigue Failure Mechanism of
Reinforced Concrete Bridge Deck Slabs,” Transportation
Research Record No. 664, Washington, D.C., 1978.
8-9. Frantz, G. C., and Breen, J. E., “Design Proposal for
Side Face Crack Control Reinforcement for Large Rein-
forced Concrete Beams,” Concrete International, V. 2, No.
10, Oct. 1980, pp. 29-34.
8-10. Grossman, J. S., “Simplified Computations for
Effective Moment of Inertia Ie and Minimum Thickness to
Avoid Deflection Computations,” ACI JOURNAL, Proceed-
ings V. 78, No. 6, Nov.-Dec. 1981, pp. 423-439.
8-11. “Design of Concrete Beams for Torsion: A Design
Aid,” PCA Bulletin EB 106.01D, 1983.
343R-102 MANUAL OF CONCRETE PRACTICE
Post-tensioned concrete bridge in California cast monolithically with an integrated pier.
CHAPTER 9—PRESTRESSED CONCRETE
9.1—Introduction
9.1.1 General—The design of prestressed concrete bridges
differs from conventionally reinforced concrete bridges in
the use of high-strength steel, generally combined with high-
strength concrete, in which stresses are induced prior to the
application of external loads. Prestressing of the steel and the
concrete improve the behavior and strength of a concrete
bridge. Prestressing the steel enables it to work at a higher
stress level. Precompressing the concrete delays its cracking.
While this active combination of the two materials should
yield desirable results, it can also impose certain problems,
by aggravating short-term dimensional changes and time-
dependent deformations of the bridge.
Most of the principles for the layout of nonprestressed
concrete bridges apply to prestressed ones. However,
because of the lighter weight of prestressed bridges and the
effects of prestressing; new concepts, techniques and
methods have been developed for them. This chapter deals
with those areas in which prestressed designs differ notably
from nonprestressed designs.
9.1.2 Codes—The considerations included herein are
essentially extracted from the AASHTO Specifications for
Highway Bridges, the ACI 318 Building Code Requirements
for Reinforced Concrete or the AREA Manual of Railway
Engineering, Chapter 8, and have been extended to permit
development of new types of concrete bridges and longer
spans. Where conflicts exist between the recommendations
of those three codes, or where those three codes are not
followed, explanations are generally provided for the
reasons underlying the recommendations of this chapter.
9.2—General design considerations
9.2.1 General—Prestressed concrete bridges should be
designed by the strength method detailed in Chapter 7 and
the service load design method detailed in Chapter 8.
9.2.2 Critical loads—Prestressed concrete bridges should
be investigated for stresses and deformations for each load
stage that may be critical during construction: stressing,
handling, transportation, and erection, as well as during the
service life of the bridge.
9.2.3 Crack control—Complete freedom from cracking
may or may not be necessary at any particular loading stage.
When cracking is permitted under service loadings, the
possibility of fatigue failure and tendon corrosion should be
investigated in accordance with the criteria recommended in
References 9-1 and 9-2.
9.2.4 Deformation stresses—Account should be taken of
the effects on adjoining structures of elastic and plastic
deformations, deflections, changes in length and rotations
caused by the prestressing. When those effects are additive
to loading, temperature, and shrinkage effects, the analysis
should consider them as acting concurrently. When the
prestress in a member affects any other member to be
designed by other individuals, the latter should be so
informed.
9.2.5 Buckling—The possibility of buckling of a slender
member or flange subjected to compressive loading, should
be considered.
9.3—Basic assumptions
Basic assumptions for prestressed concrete members for
the service load design method are specified in Section 8.1
and in Section 7.3 for the strength method.
 BRIDGE ANALYSIS AND DESIGN 343R-103

9.4—Flexure, shear
Flexure and shear strength investigations should be made
in accordance with Chapter 7.
9.5—Permissible stresses
Permissible stresses for concrete and steel at service
conditions should be those specified in Section 8.7.
9.6—Prestress losses
9.6.1 General—To accurately determine the effective
prestress, allowance should be made for the following
sources of prestress loss:
a. Slip at anchorage.
b. Friction losses due to intended and unintended curva-
ture in the tendons.
c. Elastic shortening of concrete.
d. Creep of concrete.
e. Shrinkage of concrete.
f. Relaxation of steel stress.
The amount of prestress loss due to these causes depends on
properties of the materials used in the bridge, the environment,
and the stress levels at various loading stages. Accurate esti-
mates of prestress loss require recognition that the losses
resulting from most sources listed above are interdependent.
9.6.2 Anchorage slip—Slip at the anchorage affects losses
for post-tensioned construction. The magnitude of the slip
depends on the prestressing system used and generally does
not vary with time. Calculations should be made in accor-
dance with a method (9-3 to 9-6) consistent with recognition
of the friction coefficients specified in Section 9.6.3.
9.6.3 Friction losses
9.6.3.1 Post-tensioned construction—Friction losses
should be based on the wobble and curvature coefficients
listed in Table 9.6.3.1, and should be verified during stressing
operations. Values of coefficients assumed for design and the
acceptable ranges of jacking forces and steel elongations
should be shown on the construction drawings. Losses due to
friction between the prestressing steel and the duct enclosure
should be estimated by the following equations
ff = fpo[l – e–(Kl + μα)]
When (Kl + μα) is not greater than 0.3, the following equa-
tion may be used
ff = fpo(Kl + μα) (9-1b)
where fpo is the stress at the jacking end, K and μ are wobble
and curvature coefficients listed in Table 9.6.3.1, l is the
length of duct from jacking end to point l, and α is the total
angular change of prestressing steel profile from jacking end
to point l from the end in radians.
For tendons confined to a vertical plane, α is the sum of
the absolute values of angular changes over length l. For
tendons curved in three dimensions, the total tridimensional
angular change α is obtained by adding, vectorially, the total
vertical angular change αv and the total horizontal angular
change αh; αv and αh are the sum of absolute values of
angular changes over length l of the projected tendon profile
in the vertical and horizontal planes respectively. The scalar
sum of αv and αh can be used as a first approximation to
represent α.
When the developed elevation and plan of the tendons are
parabolic or circular, the total angular change α can be
computed from
2 2
α = αv + αh (9-2a)
where
α = tridimensional angular change
αv = angular change in developed elevation
αh = angular change in plan
When the developed elevation and the plan of the tendon
are generalized curves, then the tendon can be split up into
small intervals and the previous formula can be applied to
each interval so that
2 2
α = ΣΔα = Δα v + Δα h (9-2b)
As a first approximation, the tendon may be replaced by a
series of chords connecting nodal points. The angular
changes Δαv and Δαh of each chord are obtained from its
(9-1a) slope in the developed elevation and in plan.

Table 9.6.3.1—Friction coefficients for post-tensioning tendons*

Wobble coefficient, K
Type of tendons and sheathing per ft per m Curvature coefficient H
Tendons in flexible metal sheathing 0.0010-0.0015 0.0033-0.0049 0.15-0.25
- wires
- 7-wire strands 0.0005-0.0020 0.0016-0.0066 0.15-0.25
- high-strength bars 0.0001-0.0006 0.0003-0.0020 0.08-0.30
Tendons in rigid and semi-rigid
galvanized
- 7-wire strands 0.0002 0.00066 0.15-0.25
Pregreased tendons 0.0003-0.0020 0.0010-0.0066 0.05-0.15
- wires and 7-wire strands
Mastic-coated tendons 0.0010-0.0020 0.0033-0.0066 0.05-0.15
- wires and 7-wire strands
* See also manufacturers’ literature or test data.
343R-104 MANUAL OF CONCRETE PRACTICE
Table 9.6.4.1—Elastic and time-dependent losses*
fc′ = 4000 psi (28 MPa)
Pretensioning with strand
Post-tensioning with strand 32,000 psi (221 MPa)
or wire
Post-tensioning with high 22,000 psi (152 MPa)
tensile bars
* Applies to bridges exposed to average condition with prestress levels inducing maximum stresses close to those specified in
Section 8.7.
9.6.3.2 Pretensioned construction—For deflected
prestressing tendons, account should be taken of losses
occurring at hold-down devices where appropriate.
9.6.4 Elastic and time-dependent losses
9.6.4.1 Approximation of losses—For prestressed
nonsegmental bridges with spans up to 120 ft (37 m),
constructed of normal-weight concrete with a strength of
3500 psi (24 MPa) or more at the time of prestressing, the
values given in Table 9.6.4.1, mostly taken from the
AASHTO specifications, can be used as approximate indica-
tors of elastic shortening and time-dependent losses.
9.6.4.2 Calculation of losses—For stress and strength
calculations of prestressed nonsegmental members with
spans not greater than 150 ft (46 m) and normal weight
concrete with a strength in excess of 3500 psi (24 MPa) at the
time of prestress, more realistic values of losses than those
given by Section 9.6.4.1 can be determined, as described in
the AASHTO Standard Specifications for Highway Bridges,
that is
Pretensioned construction:
Δf p = Δf e + Δf c + Δf r (9-3)
Post-tensioned construction:
Δf p = 0.5Δf e + 0.8Δf s + Δf c + Δf r (9-4)
where
Δfp = total loss in prestressing steel stress. The value of
fp does not include friction and anchorage slip
effects
Δfe = loss in prestressing steel stress due to elastic
shortening
Δfs = loss in prestressing steel stress due to shrinkage
Δfc = loss in prestressing steel stress due to creep
Δfr = loss in prestressing steel stress due to relaxation
Elastic shortening losses Δfe can be estimated from the
formula
Δfe = (Eps/Eci)fcir (9-5)
where
Eps = modulus of elasticity of prestressing strand
Eci = modulus of elasticity of concrete at transfer of
stress
fcir = concrete stress at center of gravity of prestressing
steel immediately after transfer. For bonded
fc′ = 5000 psi (35 MPa) fc′ = 6000 psi (42 MPa)
45,000 psi (310 MPa) 45,000 psi (310 MPa)
33,000 psi (228 MPa) 35,000 psi (241 MPa)
23,000 psi (159 MPa) 24,000 psi (165 MPa)
structures of usual design, the fcir value should be
at the section or, for continuous construction, the
sections of maximum moment. For unbonded
construction, the fcir value should be calculated as
the stress at the center of gravity of the
prestressing steel averaged along the length of the
member. Values of fcir should be calculated using
a steel stress reduced below the initial value by a
margin dependent on elastic shortening, relax-
ation, and friction effects. For pretensioned
members of usual design, steel stress can be
0.63fpu for tendons stressed initially to 0.7fpu. The
ultimate strength of the prestressing steel is fpu.
For post-tensioned construction, Δfe losses can be further
reduced below those implied by Eq. (9-5) with proper
tensioning procedures, such as stage stressing and reten-
sioning.
Shrinkage losses fs can be estimated from the formula
Δfs = (17000 – 150R) psi
or
(117.2 – 1.034R) MPa (9-6)
where
R = average annual ambient relative humidity,
percent.
Creep losses Δfc can be estimated from the formula
Δfc = 12fcir – 7Δfcds (9-7)
where
Δfcds = change in concrete stress at center of gravity of
prestressing steel due to all dead loads except the
dead load acting at the time the prestressing force
is applied. Values of Δfcds should be calculated at
the same section(s) for which fcir is calculated.
Losses due to relaxation of prestressing steel Δfr can be
estimated from the following
Pretensioning with 250 or 270 ksi (1725 or 1860 MPa)
stress-relieved strands:
Δfr = 20000 – 0.4Δfe – 0.2(Δfs + Δfc), psi
or
[138 – 0.4Δfe – 0.2(Δfs + Δfc), MPa (9-8)
 BRIDGE ANALYSIS AND DESIGN
Post-tensioning with 250 to 270 ksi (1725 or 1860 MPa)
stress-relieved strands
Δfr = 20000 – 0.3Δff – 0.4Δfe – 0.2(Δfs + Δfc), psi
or
[138 - 0.3Δff – 0.4Δfe – 0.2(Δfs + Δfc), MPa] (9-9)
Post-tensioning with 240 ksi (1655 MPa) stress-relieved
wire
Δfr = 18000 – 0.4Δfe – 0.3Δff – 0.2(Δfs + Δfc), psi
or
[124 – 0.4Δfe – 0.3Δff – 0.2(Δfs + Δfc), MPa] (9-10)
Post-tensioning with 145 to 160 ksi (1000 to 1103 MPa)
bars
Δfr = 3000 psi (21 MPa) (9-11)
where
Δff = friction loss below the level of 0.70
fpu computed according to Section 9.6.3.1.
Eq. (9-8), (9-9), and (9-10) are appropriate for normal
temperature ranges only. Relaxation losses increase with
increasing temperatures. For prestressing steels with low
relaxation properties (ASTM A 416 and E 328), relaxation
losses can be a quarter of those given by Eqs. (9-8), (9-9),
and (9-10).
9.6.4.3 Losses for deflection calculations—For camber
and deflection calculations of prestressed nonsegmental
members with spans up to 150 ft. (46 m) and normal-weight
concrete with a strength in excess of 3500 psi (24 MPa) at the
time of prestress, losses calculated from Eqs. (9-5) and (9-6)
should utilize values for fcir and fcds, computed as the stress
at the center of gravity of prestressing steel averaged along
the length of the member.
9.6.4.4 Loss calculations for unusual bridges—For
segmental construction, lightweight concrete construction,
stage prestressing, spans greater than 150 ft. (46 m) or for
bridges where more exact evaluation of prestress losses are
desired, calculations should be made in accordance with a
method supported by proven research data. Examples include
the time-steps method described in Reference 9-5 and 9-7 to
9-12 and the general procedure developed in Reference 9-13.
An appropriate method providing most data needed for such
calculations is detailed in Reference 9-1 and 9-14. Calcula-
tions should preferably utilize data from control tests on the
materials to be used, the methods of curing, ambient service
conditions, and pertinent structural details for the construc-
tion. Where such data are not collected, the values prescribed
in Reference 9-1 could be used.
343R-105
9.6.4.5 Effect of nonprestressed reinforcement—The
presence of a substantial amount of ordinary nonprestressed
reinforcement, such as in partial prestressing, may result in a
significant decrease in prestress losses in the steel; it may
also lead to an increase in the loss of compression in the
concrete caused by prestress losses.9-15
9.7—Combined tension and bending
Prestressed concrete tension members, whether or not
subjected to bending with or without nonprestressed rein-
forcement, can be designed by the strength method by
applying the equations of equilibrium, considering stress,
and strain compatibility and including the effects of
prestressing, shrinkage, and creep. The moment produced by
deflection and axial tension may be neglected except in very
special cases.
Prestressed concrete tension members should be investi-
gated as to stresses and deformations at various loading
stages, taking into account the effect of shortening under
prestress and recovery under external load. Typical behavior
of prestressed concrete tension members is described in
Reference 9-16, and various design approaches are devel-
oped in References 9-5, 9-6, 9-9, and 9-17.
9.8—Combined compression and bending
Prestressed concrete members under combined compres-
sion and bending, with or without nonprestressed reinforce-
ment, can be designed by the strength method based on the
same principles and assumptions for members without
prestressing by applying the equations of equilibrium,
considering stress and strain compatibility and including the
effects of prestressing, shrinkage and creep. These members
should be investigated as to stresses and deformations at
various critical loading stages, taking into account the short-
ening under prestress and external load. Members for which
the total effective prestress force divided by the gross area of
the concrete section is less than 225 psi (1.55 MPa), should
contain the minimum longitudinal reinforcement specified in
Section 7.3. For members with average prestress exceeding
225 psi (1.55 MPa), those minimum requirements can be
waived, provided analysis shows adequate strength and
stability. Except for walls or piles, all prestressing steel
should be enclosed by spirals or closed No. 3 (#10) size
lateral ties conforming to the requirements of Section 13.3.
For compression members with klu/r values exceeding the
limits specified in Section 7.3, an analysis of column
strength should be made, taking into account the effect of
additional deflection on the moments. Appropriate design
methods to account for slenderness effects are described in
Reference 9-2, 9-5, and 9-18 through 9-20.
9.9—Combination of prestressed and
nonprestressed reinforcement—Partial prestressing
Members may be reinforced by a combination of
prestressed and nonprestressed reinforcement. Such
members should be designed by the strength method, consid-
ering the stress and strain compatibility of the various steels,
including the effect of prestressing, shrinkage and creep.
343R-106 MANUAL OF CONCRETE PRACTICE
Members so designed should be checked for serviceability
requirements, particularly with regard to extent of cracking,
amount of deflection or camber at various stages of loading,
and fatigue. Design methods for service and ultimate loads,
as well as for various serviceability criteria can be found in
References 9-21 through 9-31.
9.10—Composite structures
9.10.1 General—Composite structures in which the deck
is assumed to act integrally with the beam should be inter-
connected in accordance with Section 9.10.2 through 9.10.4
to transfer shear along contact surfaces and to prevent sepa-
ration of elements.
9.10.2 Shear transfer—Full transfer of horizontal shear
forces should be assured at contact surfaces of intercon-
nected elements. Similar to the design for vertical shear
(Section 7.3), the design for horizontal shear transfer should
be based on strength requirements in accordance with the
following relation
Vu ≤ Vnh (9-12)
where Vu is the factored shear force at section considered, φ
is the strength reduction factor for shear, and Vnh is the
nominal horizontal shear strength.
9.10.3 Shear capacity—The following values of nominal
shear resistance may be assumed at the contact surface of
minimum width bv:
Vnh = 80bvd in pounds (0.552bvd in MN) when contact
surfaces are clean, free of laitance, and intentionally
roughened to a full amplitude of at least ±1/4 in.
(±6 mm).
Vnh = 80bvd in pounds (0.552bvd in MN) when minimum
ties are provided in accordance with Section
9.10.4, and the contact surfaces are clean and free
of laitance, but not intentionally roughened.
Vnh = 350bvd in pounds (2.413bvd in MN) when
minimum ties are provided in accordance with
Section 9.10.4, and the contact surfaces are clean,
free of laitance, and intentionally roughened to a
full amplitude of ±1/4 in. (±6 mm).
When the factored shear force Vu exceeds 350bvd in
pounds (2.413bvd in MN), the design should be in accor-
dance with Section 11.7 of ACI 318, which is based on eval-
uating shear friction at the interface.
9.10.4 Vertical ties—All web reinforcement should
extend into the cast-in-place decks. The minimum total area
of vertical ties per linear foot of span should not be less than
the area of two No. 3 bars spaced at 12 in. (300 mm). Web
reinforcement may be used to satisfy the vertical tie require-
ment. The spacing of vertical ties should not be greater than
four times the average thickness of the composite flange and
in no case greater than 24 in. (600 mm).
9.10.5 Shrinkage stresses—In structures with a cast-in-
place slab on precast beams, the differential shrinkage tends
to cause tensile stresses in the slab and in the bottom of the
beams. Because the tensile shrinkage develops over an
extended time period, the effect on the beams is reduced by
creep. Differential shrinkage may influence the cracking
load and the beam deflection profile. When these factors are
particularly significant the effect of differential shrinkage
should be added to the effect of loads.
9.11—Crack control
9.11.1 General—Where permissible, tensile stresses
under full service live load exceed the values recommended
in Section 8.7, or when unbonded construction is used, the
provisions of this section on crack control should be applied.
The provisions of this section are not applicable to bridges
composed of precast segments having epoxy-glued joints.
Flexural tensile stresses at joints between adjacent segments
should be limited to zero, unless nonprestressed reinforce-
ment is added to balance the tensile stresses as calculated.
9.11.2 Construction using bonded tendons—Flexural
tensile stresses for longitudinal bending only, evaluated for
full service live load and assuming an uncracked section,
should not exceed the following:
For secondary road bridges in noncorrosive environ-
ments — 12 f c ′ psi ( f c ′ MPa).
For secondary road bridges in corrosive environments
or primary road bridges in noncorrosive environments
— 6 f c ′ psi (0.5 f c ′ MPa).
For primary road bridges in corrosive environments —
3 f c ′ psi (0.25 f c ′ MPa).
When using the allowable tensile stresses previously given,
the effects of temperature gradients through the members
during service should be included in the analysis. Localized
stress concentrations should be considered. Sufficient area of
bonded steel should be placed in the area of tension to
distribute the cracks. An example of analysis for temperature
induced stresses can be found in Reference 9-32.
A primary road is one in which full service live load is
experienced regularly on an hourly basis and for which a
repair would cause a major disruption of traffic. A secondary
road is one in which traffic is infrequent and a disruption
tolerable. A corrosive environment exists in coastal areas, in
wet city smog, and in places where road salt can attack the
concrete being considered.9-33 These flexural stresses are
maximum values imposed for esthetic reasons or for protec-
tion of the reinforcement against corrosion. They could be
exceeded, provided it is shown by an accurate analysis or by
tests that performance will not be impaired.
9.11.3 Construction using unbonded tendons—In order to
control and limit cracking of beams with unbonded tendons
under occasional overloads, a minimum amount of bonded
reinforcement, whether prestressed or not, should be used as
follows
Asa = ρAg (9-13)
where Asa is the area of bonded steel (all grades), Ag is the
gross area of concrete, and ρ is given by:
ρ = 0.0025 – 10–6Ag for Ag in in.2
 BRIDGE ANALYSIS AND DESIGN
ρ = 0.0025 – 0.155 × 10–4Ag for Ag in mm2 (9-14)
The bonded reinforcement should be uniformly distrib-
uted over the precompressed tension zone as close as
possible to the extreme tension fiber.
The bonded reinforcement may be reduced to that which,
in combination with the unbonded tendons, develops a flex-
ural strength equal to that for the same tendons bonded,
provided the tensile stress in the member does not exceed
7.5 f c ′ psi (0.62 f c ′ MPa) under an occasional overload of
D + 1.2 (L + I), where I represents impact effects. Tensile
stresses from all sources should be considered; in particular,
tensile stresses arising from differential temperature distri-
butions, differential shrinkage between composites slabs and
girders, support settlement in continuous beams, and local
effects due to tendon anchorage.
9.12—Repetitive loads
The possibility of failure of steel reinforcement due to
repeated loads should be investigated in regions where flex-
ural cracking is expected at design loads.
9.12.1 Construction using unbonded tendons—In
construction with unbonded tendons subjected to repetitive
loads, special attention should be given to the possibility of
fatigue in the anchorages and couplers.9-34–9-36 Generally,
the stress range in the tendon at the anchorage is quite small
and the possibility of fatigue failure is remote unless large
deflections are repeatedly produced or the seating of the
anchorage is poor.
9.12.2 Diagonal tension—The possibility of inclined
diagonal tension cracks forming under repetitive loading at
appreciably smaller stresses than under static loading should
be considered in the design.9-34
9.12.3 Fatigue—Pertinent information on the fatigue
characteristics of prestressing steels, anchorages, and
prestressed beams are summarized in Reference 9-34. Some
additional information can be found in Reference 9-5.
9.13—End regions and laminar cracking
9.13.1 Cracking—Cracking results from the flow of
tendon stress from a concentrated area at the anchorage to
the linear distribution associated with the member. Three
types of cracking are possible: bursting or splitting, spalling,
and section change.9-36,9-37 Bursting or splitting cracks prop-
agate along the tendon axis and initiate at a distance equal to
the bearing plate width beneath the loaded surface. Spalling
cracks initiate at the end face of the member, some distance
removed from the tendon axis, and propagate parallel to the
longitudinal axis of the member. Section change cracks
result from too sharp a transition from the linear stress distri-
bution necessitated by one section shape to that necessitated
by the adjacent section shape. Section change cracks may
also be caused by bursting stresses, if the center of the tendon
anchorage is located closer than the bearing plate width to a
thinner section. Under sustained load, all three crack types
increase in width and length. Additional information on the
cracking behavior of post-tensioned girder anchorage zones
and their design can be found in References 9-38 and 9-39.
343R-107
9.13.1.1 Bursting or splitting cracks—For a single
tendon anchorage, the possibility of cracking can be readily
assessed using information given in many texts.9-6,9-8 Data
for a single centrally loaded anchorage can be extended to
multiple anchorages by assuming each anchorage to act as a
single anchorage centrally placed on an area whose semi-
axes are defined by the distance to the nearest edge or half
the distance to the adjacent anchorage.9-40 The maximum
bursting stress should not exceed 6 f ci ′ psi (0.5 f ci ′ MPa)
for an end region without reinforcement9-28 and [(6 f ci ′ ) +
3000 (Ase/bese)] psi {[(0.5 f ci ′ + 0.44 (Ase/bese)] MPa} for a
region with reinforcement. Ase is the area of each stirrup used
as reinforcement, se is the spacing of stirrups, and be the
width of the section in the plane of the potential crack.
Bursting reinforcement should extend over the length of the
bursting zone for which the computed stress exceeds 4 f ci ′
psi (0.33 f ci ′ MPa). Even where a centrally-loaded bursting
zone is well reinforced, the increment of load between the
visual observation of a bursting crack and its spontaneous
propagation is small; bursting cracks are not likely in preten-
sioned beams.
9.13.1.2 Spalling cracks—The critical location for spal-
ling cracks should be determined by finding the longitudinal
plane on which the difference in moment between that
caused by the linear stress distribution in the member and
that caused by the stress at the tendon anchorages is a
maximum. The possibility of cracking in both horizontal and
vertical planes should be considered. Stresses and the
amount of reinforcement to control cracking should be deter-
mined using the approximate method described in Refer-
ences 9-41 and 9-42 or the more exact procedure described
in Reference 9-36. If the stress caused by the difference in
moment exceeds φ 4 f c ′ psi (φ 0.33 f c ′ MPa), reinforce-
ment is necessary to control the crack growth, although spal-
ling cracks will not propagate spontaneously until their
length exceeds half the specimen depth. Reinforcement to
control spalling cracks should be placed within one-fifth of
the overall depth of the end of the member from its end, with
the end stirrup as close to the end of the beam as practicable.
In any member, the amount of spalling reinforcement should
be not less than that required to resist 4 percent of the
prestressing forces when acting at a unit stress of 20,000 psi
(138 MPa). This requirement can be waived if the required
force is less than φ 12be f c ′ (φ be f c ′ ). Spalling cracks are
likely in both post-tensioned and pretensioned beams.
9.13.1.3 Section change cracks—The possibility of
section change cracks due to too sharp a section transition
should be evaluated using the same procedure as for spalling
cracks.9-37
9.13.1.4 Reinforcement details—Reinforcement to
control cracking should consist of closed hoops, extending
over the full depth of the member and properly anchored to
develop their yield strength within the depth outside the
outermost crack. For end-region crack control, a large
number of small diameter stirrups is more effective than a
small number of large diameter stirrups.
9.13.2 End blocks—For beams with post-tensioning
tendons, end blocks are frequently needed to distribute the
343R-108 MANUAL OF CONCRETE PRACTICE
large concentrated prestressing forces. Where all tendons are
pretensioned, end blocks are generally not required and may
increase spalling stresses.9-41
9.13.3 Bearing under anchorages
9.13.3.1 Maximum stresses—Bearing stresses beneath
post-tensioning anchorages should not exceed values calcu-
lated from the formulas:9-33,9-43,9-44
a. Immediately after tendon anchoring:
fcp = 0.80fci′ A 2 ⁄ A 1 – 0.2 ≤ 1.25fci′ (9-15)
b. After allowance of prestress losses:
fcp = 0.6fc' A 2 ⁄ A 1 ≤ fc' (9-16)
where
A1 = Bearing area of anchor plate of post-tensioning
tendon
A2 = Maximum area of the anchorage surface that is
geometrically similar to and concentric with the
anchor plate and does not overlap the similar area
surrounding an adjacent anchorage
Consideration of the strength reduction factor φ is already
incorporated in Eq. (9-15) and (9-16).
9.13.3.2 Conical anchorages—Where conical-shaped
anchorages are used with surfaces inclined at angles between
70 and 15 deg to the tendon axis, allowable stresses should
be based on the maximum area of the anchorage and stresses
should be reduced 15 percent below those calculated from
Section 9.13.
9.13.3.3 Plate thickness—The thickness of the bearing
plate ha should be greater than that given by the formula9-40
ha = (2fci′/fya)A2/A1 (9-17)
where A1 and A2 have the values specified in Section 9.13,
and fya is the yield strength of the bearing plate. If a lesser
thickness of bearing plate is used, area A1 in Section 9.13.3.1
should be the area enclosed within a perimeter located a
distance ha outside the perimeter of the anchorage.
9.13.4 Inclined tendons—Large increases in spalling
stress occur for small inclinations of the tendon force.9-40
Where cracking is predicted, reinforcement sufficient to
control spalling should be provided to carry at its yield
strength, 1.3 times the component of the tendon force acting
transverse to the surface of the member.
9.13.5 Laminar cracking—In beam flanges and slabs,
laminar in-plane cracks may occur, especially in areas
having high concentrations of ducts and tendons in either
one or two directions. Laminar cracking is due to a substan-
tial reduction in the tensile area of concrete available to resist
principal tension resulting from any cause. To reduce the
development and extension of laminar cracks, it is good
practice to provide a percentage of through-slab ties of at
least 0.1 percent, with 0.2 percent preferred for severe cases.
Additional information can be found in Reference 9-45.
9.14—Continuity
9.14.1 General—Continuous beams and other statically
indeterminate structures should be designed to possess satis-
factory service load behavior and specified safety factors at
ultimate load. Behavior should be determined by elastic
analysis, taking into account the secondary reactions,
moments, shears, axial deformation, due to prestressing,
creep, shrinkage, temperature, restraint of attached structural
elements and foundation settlement.
9.14.2 Continuous bridges
9.14.2.1 General—For cast-in-place continuous bridges
or bridges made continuous by post-tensioning over two or
more spans, the effects of secondary moments due to the
reactions induced by prestressing should be included in
stress calculations for checking serviceability behavior and
for design by the service load method. For bonded and
unbonded construction, in which sufficient bonded steel is
provided at the supports to assure control of cracking
(Section 9.11), negative moments due to factored loads,
calculated by elastic theory for any assumed loading
arrangement, may be increased or decreased by not more
than (20 - 47 c/d) percent,9-46 but not more than 20 percent,
provided that these modified negative moments are also used
for the final calculations of moments at other sections in the
span for the assumed loading arrangement. In the previous
expression, the value of c is the distance from the extreme
compression fiber to the neutral axis, and d is the distance
from the extreme compression fiber to the centroid of the
tensile force in the steel.
For design by the strength method, the moments to be used
should be the algebraic sum of the moments due to the reac-
tions induced by prestressing (with a load factor of 1.0) and
the moments, including redistribution effects, due to the
service loads multiplied by the strength design load factors
specified in Chapter 5.
9.14.2.2 Minimum dead load—In the application of
strength design load factors, where the effects of dead and
live loads are of opposite sign, the case of a dead load factor
of 0.9 should be included in the investigation.
9.14.3 Bridges composed of girders made continuous
9.14.3.1 General—When structural continuity is
assumed in calculating live loads, plus impact and composite
dead load moments, the effects of creep and shrinkage
should be considered in the design of bridges incorporating
simple span precast, prestressed girders and deck slabs
continuous over two or more spans.
9.14.3.2 Positive moment connection at piers—Provi-
sion should be made in the design for the positive moments
that may develop in the negative moment region due to the
combined effects of creep, shrinkage, and temperature in the
girders and deck slab plus the effects of live load plus impact
in remote spans. Shrinkage and elastic shortening of the piers
should also be considered.
Nonprestressed positive moment connection reinforcement
at piers may be designed for a service load stress of 0.6 times
the yield strength, but not to exceed 36,000 psi (248 MPa).
 BRIDGE ANALYSIS AND DESIGN
9.14.3.3 Negative moments—Negative moment rein-
forcement should preferably be proportioned by the strength
design method.
The effect of initial precompression due to prestress in the
girders, may be neglected in the negative moment calcula-
tion if the maximum precompression stress is less than 0.4fc′,
and the continuity reinforcement ratio p in the deck slab is
less than 0.015.
With long span composite bridges at maximum girder
spacing, made continuous for live load through a cast-in-
place slab, creep and shrinkage effects may produce negative
restraining moments. These effects are to be considered not
only at the supports, but throughout the span.
9.14.3.4 Compressive stress at piers under service
loads—The compressive stress in ends of girders at interior
supports, resulting from the addition of the effects of
prestressing and service live load bending, should not exceed
0.60fc′.
9.15—Torsion
9.15.1 General—A concrete member under torsion, when
prestressed, generally possesses higher resistance against
cracking because the axial compression in the member
reduces the principal tensile stress. However, the failure of
such a member under torsion can be sudden and explosive
because of the energy stored in the member. Hence, while
prestressing can improve the elastic behavior of a concrete
member subjected to torsion, it can also cause an undesirable
mode of failure.
9.15.2 Curved bridges—The behavior of a curved bridge
under torsion can be improved by prestressing. Prestressing
can be designed to reduce the bending of the member, thus
leaving a higher reserve strength to carry the torsion. If a
simply supported beam has stiff intermediate and end
diaphragms, torsional moments can be transferred to the
supports. A bridge with redundant reactions could be pre-
stressed to reduce the torsion on the curved structure. For
example, a continuous multigirder bridge on a curve with
transverse diaphragms can be post-tensioned with more
prestress on the outer girders than the inner ones, to apply a
counter-torque which reduces the torsion on the bridge.
9.15.3 Design neglecting torsional stiffness—A capacity
in excess of that causing torsional cracking can be utilized if
torsional stiffness effects are neglected in the analysis for the
distribution of the loads to the various resisting elements of
the bridge. Adequate torsional capacity should be provided
through the addition of nonprestressed steel in the form of
ties, spirals, and longitudinal bars. Information on post-
cracking torsional behavior and reinforcement requirements
for adequate torsional resistance is contained in Ref. 9-47
through 9-50.
9.15.4 Design including torsional stiffness—Where reli-
ance is placed on torsional stiffness effects in the analysis for
the distribution of loads to the resisting elements of the
bridge, the maximum principal tensile stress evaluated by
elastic theory and for the load factors specified in Section 7.3
should not exceed φ 4 f c ′ psi (φ 0.33 f c ′ MPa). Tensile
343R-109
stresses from all sources should be considered including
those arising from differential temperature distributions.
9.16—Cover and spacing of prestressing steel
The minimum concrete cover should be provided
according to Section 13.8.
The minimum clear spacing between prestressing tendons
should satisfy the requirements of Section 13.7.
9.17—Unbonded tendons
9.17.1 General—In bridge construction, unbonded
tendons are used less frequently than bonded tendons (1)
because of the difficulty of ensuring adequate corrosion
protection, (2) because of the lowering of the ultimate
strength relative to that for bonded tendons, (3) because of
susceptibility to catastrophic failure, and (4) because wide
cracks might develop under overloads unless members are
adequately reinforced with additional bonded steel.
However, unbonded tendons can be and have been used for
prestressed suspension bridges, for box girders where the
tendons can be inspected, where safety from corrosion can
be insured and where the greater ease of construction has
resulted in a more economical structure. When unbonded
tendons are used, “Recommendations for Concrete Members
Prestressed with Unbonded Tendons,” by ACI-ASCE
Committee on Prestressed Concrete9-35,9-51 can be referred to
as a minimum standard, particularly applicable sections
which are incorporated in various sections of this report,
namely; Corrosion Protection, Section 9.17.2, and Crack
Control, Section 9.11.2.
9.17.2 Corrosion protection—To insure corrosion protec-
tion, unbonded tendons MUST be properly coated with a mate-
rial having, as nearly as possible, the following properties:
1. Free from cracks and not brittle or fluid over the entire
anticipated range of temperatures (usually taken as –40 F to
160 F) (–40 C to 70 C).
2. Chemically stable for a period of at least 50 years.
3. Nonreactive to chlorides and to material used for casing
and sheathing.
4. Noncorrosive for tendons and anchorages.
5. Impervious to moisture.
6. Adherent to tendons.
Sheaths used on unbonded tendons should prevent the
intrusion of cement paste and the escape of coating material.
They should have sufficient tensile strength and water-resis-
tance to avoid damage and deterioration during transit,
storage at job site, and installation.
The anchorage zones of bridges with unbonded tendons
should be detailed with special care to avoid deterioration, as
well as that of the surrounding concrete, due to road salts and
other corrosive agents.
9.18—Embedment of pretensioning strands
Pretensioning strands should be embedded in accordance
with Section 13.2.
343R-110 MANUAL OF CONCRETE PRACTICE
9.19—Concrete
9.19.1 Admixtures—Admixtures to obtain high-early
strength or to increase the workability of low-slump
concrete, meeting the requirements of ASTM C 494, may be
used if known to have no injurious effects on the steel or the
concrete. Calcium chloride or an admixture containing chlo-
ride ions in other than trace amounts should not be used. Sea
water should not be used as mixing water.
9.19.2 Strength—Concrete strength required at given ages
should be indicated on the construction drawings. The
strength at transfer should be adequate for the requirements
of the anchorages or of transfer through bond, as well as for
camber or deflection requirements. For pretensioned seven-
wire strand, the minimum strength at transfer should be 3000
psi (21 MPa) for 3/8-in. (9.5-mm) strands and smaller, and
3500 psi (24 MPa) for 7/16-in. (11.1-mm) and 1/2-in.
(12.7-mm) strands. AASHTO recommends the use of a
minimum strength at transfer of 4000 psi (28 MPa) for
pretensioned members and 3500 psi (24 MPa) for post-
tensioned members.
9.20—Joints and bearings for precast members
9.20.1 General—Design and detailing of the joints and
bearing should be based on the forces to be transmitted, and
on the effects of dimensional changes due to shrinkage,
elastic deformation, creep, and temperature. Joints should be
detailed to allow sufficient tolerances for manufacture and
erection of the members. Permissible ultimate bearing
stresses for concrete increase with the confining effects of
surrounding concrete and reinforcement, and decrease as the
stiffness of the loading plate decreases or the eccentricity of
the load area relative to the centroid of the resisting area
increases. Bearing stresses should not exceed the values
specified in Section 7.3.
9.20.2 Design criteria—Fixed bearings should be detailed
to provide for stress concentrations, rotations, and the
possible development of horizontal forces by friction,
shrinkage, and creep, or other restraints. Expansion bearings
should be detailed to provide for stress concentrations, rota-
tions, and 125 percent of the horizontal movements due to
elastic shortening, creep, shrinkage, and temperature effects.
Desirable characteristics for bearings are specified in AASHTO
1983, Sections 1.5.23 (H) and 1.7.32 through 1.7.38.
9.21—Curved box girders
In curved prestressed box girder construction, the radial
tendon forces which are normal to the web plane should be
taken into consideration. Local tendon-induced bending,
shear, and the force transfer from tendon to flanges should be
considered in the design and detailing.
RECOMMENDED REFERENCES
The documents of the various standards-producing organi-
zations referred to in this report are listed below with their
serial designation, including year of adoption or revision.
The documents listed were the latest effort at the time this
report was written. Since some of these documents are
revised frequently, generally in minor detail only, the user of
this report should check directly with the sponsoring group
if it is desired to refer to the latest revision.
American Association of State Highway and Transportation
Officials
Standard Specifications for Highway Bridges, Thirteenth
Edition, 1983.
American Concrete Institute
318-83 Building Code Requirements for Rein-
forced Concrete
ACI-ASCE 423 Recommendations for Concrete
Members Prestressed with Unbonded
Tendons
American Railway Engineering Association
Chapter 8, Manual for Railway Engineering
ASTM International
A 416-85 Specification for Uncoated Seven-Wire Stress-
Relieved Steel Strand for Prestressed Concrete
C 494-82 Specification for Chemical Admixtures for
Concrete
E 328-78 Recommended Practice for Stress-Relaxation
Tests for Materials and Structures
CITED REFERENCES
9-1. PCI Committee on Prestress Losses, “Recommenda-
tions for Estimating Prestress Losses,” PCI Journal, V. 20,
No. 4, July-Aug. 1975.
9-2. Nathan, N. D., “Applicability of ACI Slenderness
Computations to Prestressed Concrete Sections,” PCI
Journal, V. 20, No. 3, May-June 1975, pp. 68-85.
9-3. Huang, T., “Anchorage Take-Up Loss in Post-
Tensioned Members,” PCI Journal, V. 14, No. 4, Aug. 1969,
pp. 30-35.
9-4. State of California, Division of Highways, Manual of
Bridge Design Practice, Third Edition, Sacramento, CA,
1983.
9-5. Naaman, A. E., Prestressed Concrete Analysis and
Design Fundamentals, McGraw-Hill Book Co., New York,
1982, 670 pp.
9-6. Lin, T. Y., and Burns, N. H., Design of Prestressed
Concrete Structures, Third Edition, John Wiley and Sons,
New York, 1981, 646 pp.
9-7. Subcommittee 5, ACI Committee 435, “Deflection of
Prestressed Concrete,” Prestressed Concrete Institute, V. 60,
No. 12, Dec. 1963, pp. 1967-1727.
9-8. Libby, J. R., Modern Prestressed Concrete: Design
Principles and Construction Methods, Third Edition, Van
Nostrand Reinhold Co., New York, 1984, 629 pp.
9-9. Nilson, A. H., Design of Prestressed Concrete, John
Wiley and Sons, New York, 1978, 526 pp.
9-10. Tadros, M. K.; Ghali, A.; and Dilger, W. H., “Time-
Dependent Analysis of Composite Frames,” Journal of the
Structural Division, ASCE, V. 103, No. ST4, April 1977, pp.
871-884.
9-11. Tadros, M. K.; Ghali, A.; and Dilger, W. H., “Long-
 BRIDGE ANALYSIS AND DESIGN
Term Stresses and Deformations of Segmental Bridges,”
PCI Journal, V. 24, No. 4, Jul.-Aug. 1979, pp. 66-87.
9-12. Scordelis, A., “Analytical Models for Nonlinear
Material, Geometric, and Time-Dependent Effects,” in
Proceedings, International Symposium on Nonlinearity and
Continuity in Prestressed Concrete, M. Z. Cohn, ed., V. 2,
University of Waterloo, Canada, June 1983, pp. 25-43.
9-13. Huang, T., “Study of Prestressed Losses Conducted
by LeHigh University,” PCI Journal, V. 27, No. 5, Sept.-
Oct. 1982, pp. 48-61.
9-14. Zia, P.; Preston, H. K.; Scott, N. L.; and Workman,
E. B., “Estimating Prestress Losses,” Concrete Interna-
tional, V. 1, No. 6, June 1979, pp. 32-38.
9-15. Tadros, M. K.; Ghali, A.; and Dilger, W. H., “Effects
of Nonprestressed Steel on Prestress Loss and Deflection,”
PCI Journal, V. 22, No. 2, Mar.-Apr. 1977, pp. 50-63.
9-16. Wheen, E. J., “Prestressed Concrete Members in
Direct Tension,” Journal of the Structural Division, ASCE,
V. 105, No. ST7, July 1973, pp. 1471-1487.
9-17 Naaman, A. E., “Optimum Design of Prestressed
Tension Members,” Journal of the Structural Division,
ASCE, V. 108, No. ST8, Aug. 1982.
9-18. PCI Committee on Prestressed Concrete Columns.
“Recommended Practice for the Design of Prestressed
Concrete Columns and Bearing Walls,” ACI JOURNAL,
Proceedings V. 21, No. 6, Nov.-Dec., 1976, pp. 16-45.
9-19. Alcock, W. J., and Nathan, N. D., “Moment Magni-
fication Tests of Prestressed Concrete Columns,” PCI
Journal, V. 22, No. 4, Jul.-Aug. 1977, pp. 50-61.
9-20. Nathan, N. D., “Slenderness of Prestressed Concrete
Columns,” PCI Journal, V. 28, No. 2, Mar.-Apr. 1983, pp.
50-77.
9-21. Naaman, A. E., and Siriaksorn, A., “Serviceability
Based Design of Partially Prestressed Beams—Parts 1 and
2,” PCI Journal, V. 24, No. 2, Mar.-Apr. 1979, pp. 64-89 and
V. 24, No. 3, May-June 1979, pp. 40-60.
9-22. Naaman, A. E., “A Proposal to Extend Some Code
Provisions on Reinforcement to Partial Prestressing,” PCI
Journal, V. 26, No. 2., Mar.-Apr. 1981, pp. 74-91.
9-23. Collins, N. P., and Mitchell, D., “Shear and Torsion
Design of Prestressed and Nonprestressed Concrete Beams,”
PCI Journal, V. 25, No. 5, Sept.-Oct. 1980, pp. 32-100.
9-24. Nawy, E. G., and Huang, P. J., “Crack and Deflec-
tion Control of Pretensioned Prestressed Beams,” PCI
Journal, V. 22, No. 3, May-June 1977, pp. 30-43.
9-25. Branson, D., and Trost, H., “Application of the I-
Effective Method to Partially Prestressed Member Deflec-
tion,” PCI Journal, V. 27, No. 5, Sept.-Oct. 1982, pp. 62-77.
(See also discussion in V. 28, No. 6, Nov.-Dec. 1983, pp.
131-136.)
9-26. Menn, C., “Partial Prestressing from the Designer's
Point of View,” Concrete International, V. 5, No. 3, Mar.
1983, pp. 52-59.
9-27. Naaman, A. E., “Time-Dependent Deflection of
Prestressed Beams by the Pressure-Line Method,” PCI
Journal, V. 28, No. 2, Mar.-Apr. 1983, pp. 98-119.
9-28. Huber, A. W., “Practical Design of Partially
Prestressed Concrete Beams,” Concrete International, V. 5,
343R-111
No. 4, Apr. 1983, pp. 49-54.
9-29. Cohn, M. Z., ed., Nonlinearity and Continuity in
Prestressed Concrete, Proceedings of an International
Symposium, University of Waterloo, Canada, July 4-6,
1983.
9-30. Tadros, M. K., “Expedient Service Load Analysis of
Cracked Prestressed Concrete Sections,” PCI Journal, V.
27, No. 6, Nov.-Dec. 1982, pp. 86-111. (See also discussion
V. 28, No. 6, Nov.-Dec. 1983, pp. 137-158.)
9-31. Cohn, M. Z., ed., Partial Prestressing: From Theory
to Practice, V. 1 and 2, Martinus Nijhoff Publishers,
Dordrecht, The Netherlands, 1986.
9-32. Elbadry, M. M., and Ghali, A., “Temperature Varia-
tions in Concrete Bridges,” Journal of the Structural Divi-
sion, ASCE, V. 109, No. 10, Oct. 1983, pp. 2355-2374.
9-33. ACI Committee 201, “Guide to Durable Concrete,”
ACI JOURNAL, Proceedings V. 74, No. 12, Dec. 1977, pp.
573-609.
9-34. ACI Committee 215, “Guide for Design of Concrete
Structures Subjected to Fatigue Loading,” ACI JOURNAL,
Proceedings V. 71, Mar. 1974, pp. 97-121.
9-35. ACI-ASCE Committee 423, “Tentative Recommen-
dations for Concrete Members Prestressed with Unbonded
Tendons,” ACI JOURNAL, Proceedings, V. 66, No. 2, Feb.
1969, pp. 81-87.
9-36. Welsh, W. A., Jr. and Sozen, M. A., “Investigation
of Prestressed Reinforced Concrete for Highway Bridges,
Part V: Analysis and Control of Anchorage Zone Cracking
in Prestressed Concrete,” Engineering Experiment Station
Bulletin, No. 497, University of Illinois, Urbana, IL, 1968.
9-37. Hawkins, N. M., “The Behavior and Design of End
Blocks for Prestressed Concrete Beams,” Civil Engineering
Transactions, Institution of Engineers, Australia, Oct. 1966.
9-38. Stone, W. C., and Breen, J. E., “Behavior of Post-
Tensioned Girder Anchorage Zones,” PCI Journal, V. 29,
No. 1, Jan.-Feb. 1984, pp. 64-109.
9-39. “Design of Post-Tensioned Girder Anchorage
Zones,” PCI Journal, V. 29, No. 2, Mar.-Apr. 1984, pp. 28-
61.
9-40. Guyon, Y., Prestressed Concrete, V. 1, Wiley &
Sons, New York, 1953, pp. 127-212.
9-41. Gergely, P., and Sozen, M. A., “Design of Anchorage
Zone Reinforcement in Prestressed Concrete Beams,” PCI
Proposed Revision of ACI 318-63, “Building Code Require-
ments for Reinforced Concrete,” ACI JOURNAL, Proceedings
V. 67, No. 9, Sept. 1970, pp. 700-701.
9-42. Marshall, W. T., and Mattock, A. H., “Control of Hori-
zontal Cracking in the Ends of Pretensioned Prestressed
Concrete Girders,” PCI Journal, V. 7, No. 5, Oct. 1962, pp. 56-
74.
9-43. Hawkins, N. M., “Discussion of Proposed Revision
of ACI 318-63, Building Code Requirements for Reinforced
Concrete,” ACI JOURNAL, Proceedings V. 67, No. 9, Sept.
1970, pp. 700-701.
9-44. Hawkins, N. M., “The Bearing Strength of Concrete
Loaded Through Flexible Plates,” Magazine of Concrete
Research, V. 20, No. 63, June 1968.
9-45. Gerwick, Ben C., Jr., “Causes and Prevention of
343R-112 MANUAL OF CONCRETE PRACTICE
Problems in Large-Scale Prestressed Concrete Construc-
tion,” PCI Journal, V. 27, No. 3, May-June 1982, pp. 59-75.
9-46. Naaman, A. E., “Partially Prestressed Concrete:
Review and Recommendations,” PCI Journal, V. 30, No. 6,
Nov.-Dec. 1985, pp. 30-71.
9-47. Wyss, A. N., and Mattock, A. N., “Full Scale
Torsion, Shear, and Bending Tests of Prestressed Girders,”
PCI Journal, V. 23, No. 2, Mar.-Apr. 1978, pp. 22-41.
9-48. McGee, D., and Zia, P., “Prestressed Concrete
Under Torsion, Shear, and Bending,” ACI JOURNAL,
Proceedings V. 73, No. 1, Jan. 1976.
9-49. Zia, P., and Hsu, T. T. C., “Design for Torsion and
Shear in Prestressed Concrete,” ASCE Preprint 3424, ASCE
Convention, Chicago, Oct. 1978, 17 pp.
9-50. Mitchell, D., and Collins, M. P., “Detailing for
Torsion,” ACI JOURNAL, Proceedings V. 73, No. 9, Sept.
1976, pp. 506-511.
9-51. ACI-ASCE Committee 423, “Recommendations
for Concrete Members Prestressed with Unbonded
Tendons,” Concrete International, V. 5, No. 7, July 1983,
pp. 61-76.
 BRIDGE ANALYSIS AND DESIGN
Completed Pine Valley Bridge in California post-tensioned segmental construction.
CHAPTER 10—SUPERSTRUCTURE SYSTEMS AND ELEMENTS
10.1—Introduction
The type of concrete bridge superstructure to be used can be
chosen from a wide variety, and the most suitable choice
depends on the geometry of the bridge, the span length, and the
method of construction. This chapter describes possible bridge
types and discusses their suitability for various situations.
Analysis of the superstructure to determine the displace-
ments, the internal forces, and the reactions may be done
using one of various techniques available to the designer.
The use of a computer greatly simplifies this task and makes
it possible to employ techniques for the analysis which
would have been otherwise impractical. The methods of
analysis available and the idealized structural models which
can be employed for bridges are discussed. Empirical equa-
tions which may be used to determine the bending moment
and shear in slabs and beams of bridge structures of common
types and simple geometry are presented.
10.2—Superstructure structural types
10.2.1 Nonprestressed concrete slab bridges
10.2.1.1 General—In this simplest type of bridge super-
structure, the deck slab also serves as the principal load
carrying element. The concrete slab, which may be solid,
cored, or ribbed, is supported directly on the substructure.
10.2.1.2 Cast-in-place—A cast-in-place solid slab may
be used for spans below 40 ft (12 m). For spans over 40 ft
(12 m), voided slabs become more economical by saving
weight and materials; circular void forms are normally used.
Cast-in-place slabs may be simple spans or continuous
spans.
343R-113
10.2.1.3 Precast—Precast slab decks usually consist of
a number of precast voided slab sections placed side by side.
The individual slab units are made to act integrally, either by
the use of longitudinal shear key or by casting a concrete slab
on top. Precast slabs are usually used for simple spans of less
than 50 ft (15 m). Longer spans can be achieved using box
sections.
10.2.2 Nonprestressed concrete girder bridges—This
type of superstructure consists of a deck slab supported by
nonprestressed concrete longitudinal girders. The two most
common forms are T-beams and box girders. They are
normally cast-in-place concrete, as large precast units are
usually limited to prestressed concrete.
The spans may be simply supported or continuous. Simple
spans are generally prismatic, but continuous spans may be
haunched. Diaphragms are normally provided at piers and
abutments. Intermediate diaphragms, particularly at
midspan, may be used to improve load distribution.10-1
The analysis should take into consideration the interac-
tion, where appropriate, of the axial, flexural, and torsional
stiffnesses of the longitudinal girders and transverse compo-
nents (the transverse components consist of the deck slab
and, if present, floor beams and diaphragms). This is partic-
ularly important in the cast of end spans with large skews,
especially for multigirder bridges, since the magnitudes of
reactions are significantly affected.
10.2.3 Prestressed concrete slab bridges
10.2.3.1 General—Prestressed, cast-in-place, post-
tensioned concrete slab bridges are particularly suitable for
expressway interchanges, where complex geometrics are
343R-114 MANUAL OF CONCRETE PRACTICE
frequently required. The inherent torsional stiffness of slabs
makes them suitable for curved continuous structures and
structures with single column intermediate supports. The
prestressed slab can be used over a wide range of spans from
solid slabs in the short spans, through round voided slabs in
the intermediate range, to rectangular voided slabs in the
longer range.
10.2.3.2 Cast-in-place—Cast-in-place prestressed slabs
may be used for simple spans, but are more commonly used
for continuous spans. For post-tensioned voided slabs, the
section should be solid over the piers and the abutment bear-
ings. Transverse post-tensioning of the solid sections at piers
with single supports is usually required. This form of
construction lends itself to round column piers which can
often overcome skew problems in complex interchange
layouts. At fixed piers, the slabs are generally cast integral
with the piers. At expansion piers, sliding pot bearings are
used. Continuous multispan decks can be used to reduce the
number of deck expansion joints, thereby minimizing main-
tenance problems. Such structures on horizontal curves or
single column supports require particular attention to
torsional stresses.10-2,ACI SP-35 Radial movement in horizon-
tally-curved decks may be utilized to reduce or eliminate
expansion joints.10-3
In the past, post-tensioned slabs using rigid round void
forms have sometimes exhibited longitudinal cracking over
the voids.10-3 This has been effectively controlled by using
light transverse post-tensioning throughout the length of the
bridge, and limiting the longitudinal prestress force to an
average value on the cross section of 0.16fc′.
10.2.3.3 Precast—Precast pretensioned concrete slabs
are used in many of the same applications as precast non-
prestressed concrete slabs. They are usually limited to
straight square spans and are not adaptable to varied deck
geometries. Simple spans may be made partially-continuous
for live load in the same way as precast beams or box girders.
10.2.4 Prestressed concrete girder bridges
10.2.4.1 General—Prestressed concrete girder bridges
cover a wide range of spans and types. In the short span
range, precast AASHTO beams with a composite cast-in-
place nonprestressed concrete slab are frequently used for
simple spans. A similar form of construction is used for
partially-continuous spans using I-girders and box girders in
the medium span range. In the medium to long span range,
continuous precast segmental box girders are common,
while the longest spans are generally cast-in-place segmental
box girders.
10.2.4.2 Cast-in-place—For cast-in-place construction,
the girders and slab are generally formed together and both
cast before formwork and supports are removed. This
construction is fully composite for dead load and live load.
The usual cross sections are T-beams and box girders. Spans
are usually continuous, and transverse post-tensioning of the
slab is frequently prescribed to allow the use of thinner slabs
or a reduced number of longitudinal girders at a larger
spacing. Since longitudinal post-tensioning is required on
site, transverse post-tensioning is usually economical and
normally used.
The design and analysis items given for reinforced
concrete girder bridges also apply to prestressed girder
bridges. For the box girder sections, a detailed transverse
live load analysis of the section should be carried out.
Temperature effects are important for box girders, due to the
possibility of large differential temperatures between the top
and bottom slabs.10-5
For cast-in-place segmental construction built by the
balanced cantilever method, a knowledge of the exact
construction loads is necessary in order to calculate stresses
and deformations at each stage. A knowledge of the creep
characteristics of the concrete is essential for calculating
deformations after the addition of each segment, and also to
calculate the redistribution of moments after completion and
final stressing.10-1,10-6
10.2.4.3 Precast—Standard precast prestressed beams
cover spans up to the 140 ft (43 m) range. After the beams
are erected, forms for the slabs are placed between the
beams, and a reinforced concrete slab cast in place. The slab
and beams act compositely for superimposed dead load and
live load. Intermediate diaphragms are not normally used,
and the design and analysis items given for reinforced
concrete girder bridges, Section 10.2, also apply to
prestressed multi-beam type bridges.
Precast prestressed beams can be made partially contin-
uous for multi-span bridges. This system is not only structur-
ally efficient, but has the advantage of reducing the number
of deck joints. Support moments are developed due to super-
imposed dead load, live load, differential temperature,
shrinkage, and creep.10-6 Continuity for superimposed dead
load and for live load can be achieved by casting diaphragms
at the time the deck concrete is placed. Reinforcing steel
placed longitudinally in the deck slab across the intermediate
piers will resist the tension from negative moments at the
supports. At the diaphragms, the bottom flanges of adjacent
beams should be connected to resist the tensile stress due to
positive moments generated by differential temperature,
shrinkage, and creep. Continuous spans, beyond the range of
the typical precast girders, normally can be achieved using
segmental precast pretensioned girders. These girders,
temporarily supported on bents, with joints near points of
minimum moment, are post-tensioned for continuity after
placement of the deck slab. The maximum lengths of
segments are usually determined by shipping length and
weight restrictions.
The forms of construction previously described using I-
girders can also be utilized for precast box girders. This form
of construction is useful when the superstructure construc-
tion depth is minimized due to clearance requirements. For
short and medium spans, closed boxes with precast top slabs
are used, but for long spans, trapezoidal open boxes, with the
deck cast in place, are often more economical. This method
reduces the weight for handling and shipping.
Precast segmental construction employs single or multiple
cell boxes with transverse segments post-tensioned together
longitudinally. For medium spans, the segments may be
erected for the full span on falsework before post-tensioning.
Longer spans are usually erected by the balanced cantilever
 BRIDGE ANALYSIS AND DESIGN
method, where each segment is successively stressed after
erection. The design and analysis considerations given for
cast-in-place segmental construction also apply to precast
segmental construction. The deformation of the structure
during cantilever erection is dependent upon the time differ-
ence between segment precasting and erecting. The design
calculations may need to be repeated if the construction
schedule differs from that assumed at the design stage.
Elastic analysis and beam theory are usually used in the
design of segmental box girder structures. For box girders of
unusual proportions, other methods of analysis which
consider shear lag should be used to determine the portion of
the cross section effective in resisting longitudinal
bending.10-7—10-9 Shear keys should be provided in webs of
precast segments to transfer shear forces during erection.
Possible reverse shearing stresses in the shear keys should be
investigated, particularly in segments near a pier. At the time
of erection, the shear stress carried by the concrete section
engaged by the shear key should not exceed 2 f c ′ (0.17 f c ′ ),
unless a more detailed analysis is made in accordance with
Section 7.3. The design and construction of segmental bridges
is discussed in detail in literature.10-10—10-14
10.2.5 Rigid frame bridges
10.2.5.1 General—A rigid frame is a structure
consisting of a continuous longitudinal member rigidly
connected with the vertical or inclined members which
support it. A “rigid connection” is a connection designed to
resist and transfer bending moments, shears, and axial forces
without relative displacement among the ends of various
members meeting at the connection.
Each vertical or inclined member of the frame should be
connected to the foundation in a manner that will prevent
horizontal movements at the base. Rigid frames should be
considered free to sway longitudinally due to the application
of loads, unless prevented from movement by specifically
designed external restraints.
For the analysis of the frame, column bases should be
assumed to be pin-ended or partially fixed, unless the char-
acteristics of the foundations permit them to be designed and
constructed as fully fixed. When a pin-end condition is
assumed for the analysis of the superstructure, the base of
column, footing, and piling should be designed to resist the
moment resulting from an assumed degree of restraint,
compatible with the column to footing connection, type of
footing, and the characteristics of the foundation material.
The influence of longitudinal forces, changes in tempera-
ture, shrinkage, and the yielding, horizontally and vertically,
of foundations can be of critical importance. It is necessary
to give proper attention to important changes in the moment
of inertia of the frame elements, as they may be cracked or
noncracked. In determining the elastic response of the frame,
the moment of inertia of the entire superstructure cross
section and that of the full cross section of the bent or pier
should be used.
10.2.5.2 Types of rigid frames—For a solid slab struc-
ture, the slab may be of uniform thickness throughout. Alter-
natively, the slab thickness may be increased at the supports
343R-115
by the introduction of haunches, or the soffit of the slab may
be curved or be formed of a number of straight segments.
T-beam or box girder cross sections reduce the dead load
bending moments throughout the structure, but are only
economical for longer spans.
In general, skew frames should be analyzed in three
dimensions. The finite strip or finite element methods are
recommended for this analysis. For preliminary design
purposes, the frame may be treated as a right angled frame
with the calculated moments and thrusts multiplied by a
factor equal to the square of the secant of the skew angle.
The reinforcement at the obtuse corners of skew frames
should receive special consideration. A skew angle greater
than 50 deg (0.87 rad) is impractical for rigid frame-type
bridges.
10.2.6 Arch bridges
10.2.6.1 General—Arch spans are not suitable for most
bridge locations, due to geometry restrictions or lack of a firm
foundation. Fixed arch spans should not be considered unless
the foundation’s conditions, which prevail at the site, provide
the required resistance to the arch reactions. At sites where
falsework may be difficult to construct, precast segmental
construction using tie-backs should be considered.
On spans less than 100 ft (30 m), the arch geometry should
be such that the axis of the ring conforms to either the equi-
librium polygon for full dead load, or the equilibrium
polygon for full dead load, plus one-half live load over the
full span, whichever produces the smallest bending stresses.
On short spans the deflection of the arch axis is small and the
secondary moments (represented by the thrust times the
deflection) may be neglected in the arch design.
On spans over 100 ft (30 m), it is imperative that the arch
centroidal axis coincides with the equilibrium polygon for
full dead load. Any live load system will disturb the above
equilibrium polygon and cause an elastic deflection of the
rib. This deflection creates additional moments which cause
further deflection, thus creating still more moments. If the
rib is too flexible, the arch ring may not be able to regain
equilibrium and will fail radially by buckling. To prevent
this type of failure, the arch ring should be designed for the
ultimate load and moments including the elastic and deflec-
tion effects. For a complete analysis, consideration should be
given to elastic shortening, creep, shrinkage, temperature,
variation in modulus of elasticity, and all other effects that
may change the length of the arch rib.
Arch ribs should be investigated for resistance against
transverse buckling. For buckling in the transverse direction,
it may be assumed that the arch rib is a straight column with
a length equal to the span and an axial load equal to the hori-
zontal thrust. The resistance to buckling can be based on
column design criteria.
10.2.6.2 Types—The spandrel arch may be either the
open spandrel column-type or the filled spandrel-type. In
lieu of a complete analysis of the entire structure, an open
spandrel arch may be designed for concentrated loads
applied at the points of spandrel support. The amount of live
load concentration at each point should be determined from
the live load on the roadway deck, placed to produce
343R-116 MANUAL OF CONCRETE PRACTICE
maximum stress at the section under consideration. In filled
spandrel construction, the equivalent uniform live load or
distributed axle load may be used for the arch design.
The three-hinged arch is statically determinate and is
recommended for short spans. The hinges may be reinforced
concrete or steel-pinned shoes. Concrete hinges are estheti-
cally preferable, since they are hardly noticeable compared
to steel hinges. Full fixity at the foundations may be assumed
if these are sufficiently rigid and the dead load reaction
produces a soil pressure as uniform as possible at the base of
the foundations.
10.2.7 Truss bridges—Vierendeel trusses are preferred to
triangular trusses in order to avoid congestion of reinforce-
ment at joints. Vierendeel trusses may have parallel chords
or inclined upper chords in through bridges. Because of their
greater depth, the deflection of Vierendeel trusses is usually
smaller than that of girder bridges of similar span length.
Because of the smaller bending stresses, inclined end posts
are preferred over upright end posts. Panels having lengths
greater than the truss height are generally more economical
than square or upright arrangements. In the analysis of Vier-
endeel trusses, the entire structure should be considered a
rigid frame to determine the axial, flexural, and shear loads
in each member. Vierendeel trusses with upper or lower
lateral systems should be analyzed as rigid space frames.
Members of Vierendeel trusses may be precast in a plant
and field-jointed. The availability of high-strength concrete,
high-strength reinforcing bars, and prestressing strands
makes this type of bridge adaptable to moderately long
spans, but it is not a common form of construction.
10.2.8 Cable-stayed bridges
10.2.8.1 General—Cable-stayed bridges are not
economical for spans below 500 ft (150 m). The cable-stayed
span is characterized by the straight, inclined cables which
support the deck system at one or several locations.10-15
Short spans may be analyzed by the linear elastic theory.
Cable-stayed bridges may behave nonlinearly for two
reasons: (a) the stiffness of the cables is dependent upon their
tension and their length, which in turn affects the sag of the
cable; (b) the displacements of the structure may be too large
to be ignored in the analysis. Successive approximations can
be used for the analysis. The displacements are first calcu-
lated by a linear analysis using the original geometry. This is
followed by a new linear analysis using deformed geometry
based on the results of the first analysis. This is repeated until
the displacements do not change appreciably. Three or four
iterations are usually sufficient.10-14
Cable-stayed bridges generally have satisfactory aerody-
namic characteristics, but the suspended spans should be
designed to minimize the generation of vertical and torsional
oscillations. The aerodynamic stability may need to be
checked by wind tunnel model tests.
10.2.8.2 Stiffening system—The stiffening system and
the cable supporting system share loads, depending on their
relative stiffness and the amount of live load superimposed
on the bridge. Hence, the stiffness factors should be carefully
determined to avoid errors in the stress determination. Under
dead load and median temperature, the stiffening system
may be assumed to be supported by the bearings and cable
supports. Live load and effects of change in temperature are
resisted by nonyielding supports at the bearings and yielding
supports at the cable connections. The added longitudinal
compression in the stiffening element due to the horizontal
component of the force in the stays should be considered and
can be utilized as a prestressing force.
10.2.8.3 Towers—The towers supporting the stay cables
may be one of the following types:
a. Twin vertical or inclined legs with one or more cross
struts to provide lateral stability. This type is required
where two planes of cables are used.
b. Twin inclined legs meeting at a common apex. This type
is used for single plane cable systems or where a double
cable system is splayed horizontally from the tower.
c. A single vertical leg tower, used only for single plane
cable systems.
For any of the above types, the towers may be hinged or
fixed at the base. Fixed bases are more common because
temporary fixing systems are not needed during erection. On
the other hand, a fixed base produces substantial longitudinal
bending moments in the tower legs due to longitudinal
movements of the top of the tower.
10.2.8.4 Cable systems—The stay cable system may be
in single or double planes. A variety of fan or harp configu-
rations may be used.10-16
The individual cables may be of the following types:
a. Parallel wire construction. A series of parallel wires
make up the circular-shaped cross section of the cable.
b. Parallel strand, closed construction. A series of helically
wound strands, either singly or, more commonly, in
parallel, make up the circular shape of the cable cross
section.
In cable-stayed spans where the suspension system is
virtually straight, concrete encasement of the cable should be
considered. This encasement serves to protect the cable from
corrosion, and also to reduce live load deflections, hence
providing added safety against fatigue stresses at the end
connections. However, to fully realize these benefits, the
cable should be pretensioned prior to the concrete encase-
ment to such a level that tensile stress will not develop in the
concrete under the full combined dead and live loads.
The design of the cable-stay system should take into
account not only the axial dead load, live load, and tempera-
ture stresses, but also the flexural stresses in the cable which
could initiate fatigue failure. These flexural stresses normally
originate at points of curvature change and movement.
10.2.9 Suspension bridges
10.2.9.1 General—Suspension bridges with concrete
towers, decks, and stiffening systems are less common than
cable-stayed bridges, but may be economical for spans over
500 ft (150 m) in special circumstances.10-16,10-17 The
suspension bridge has vertical hangers connecting the
suspension cables to the stiffening system. The cables may
be self-anchored to the deck or supported by independent
concrete anchorages.
 BRIDGE ANALYSIS AND DESIGN
Deflections for similar span lengths are greater than for
cable-stayed bridges. Suspension bridges should always be
analyzed by nonlinear deflection theory methods.
Suspension spans should be designed to minimize the
generation of vertical and torsional oscillations from wind,
and aerodynamic stability should be checked by wind tunnel
model tests.
10.2.9.2 Stiffening system—The stiffening system
distributes the live load to the hangers and is stressed prima-
rily by live load, temperature, and wind. The dead load stress
is small, resulting only from spanning the short distance
between hangers.
The stiffening system and deck are normally combined by
the use of box girders, or solid or voided concrete slabs. The
stiffening system may be constructed by the segmental
method. If the cable system is self-anchored, the added
longitudinal compression in the stiffening system and deck
should be considered.
10.2.9.3 Towers—There are always two cables, there-
fore, the towers are the twin leg-type with one or more cross
struts, and are usually fixed at the base.
10.2.9.4 Suspension system—The main cables may be of
parallel wire, parallel strand, or helical strand construction.
The helical strands should be prestretched before erection.
The strands usually are formed into circular cables, but may
be open-type, where the strands are separated into an open-
box-like shape by means of spacers. All wires should be
galvanized.
The hangers are usually galvanized wire ropes or bridge
strands socketed at both ends.
10.3—Methods of superstructure analysis
10.3.1 General—The analysis of bridge superstructures is
normally based on elastic or model analysis methods. All
sections should be proportioned to resist the forces deter-
mined from the results of the analysis. The analysis should
recognize stresses occurring from temperature, shrinkage,
creep, method of construction, transport and handling of
units, prestress, and other conditions which may affect the
design.10-1 Empirical methods may be employed for super-
structures of common types and simple geometry as speci-
fied in Sections 10.4 and 5.
10.3.2 Elastic methods—The forces present in the compo-
nent parts of the superstructure should normally be estab-
lished in accordance with a recognized elastic analysis,
except when an empirical method of analysis
applies.10-2,10-8,10-9,10-18—10-20 and ACI SP35 The available
elastic analytical methods considered suitable for bridge
analysis are:
a. Simplified methods. The beam analogy, in which the
superstructure is analyzed assuming it to be a beam for
evaluating longitudinal moments and shears, should be
restricted to simple structures. For structures of vari-
able width, substantial horizontal curvature, or with
skews exceeding 20 deg (0.35 rad), this method should
be used for preliminary design only.
b. Refined methods. The following general methods can
be applied to more complex structures: grillage
343R-117
analogy, folded plate, orthotropic plate theory, finite
strip, and finite element methods.10-7—10-9,10-19,10-20
The most suitable method of analysis should be selected
for the solution of each particular problem. The accuracy of
the results obtained depends on how closely the assumptions
inherent in the method of analysis conform to the actual
structure.
10.3.3 Model analysis—If analytical methods cannot be
used or if adequate facilities are readily available, model
testing should be considered. Model testing may be used to
advantage when skew or irregularly shaped superstructures
are required.10-21 and ACI SP-24 The modeling material may be
plastic, microconcrete, or other material which adequately
approximates the behavior of the prototype. The effect of
scale should be considered.
10.3.4 Nonlinear methods—Suitable nonlinear methods
are not generally used for analysis of the usual types of struc-
ture. Special structural types such as cable stayed and
suspension bridges, however, may require nonlinear
methods in order to give satisfactory results.
10.4—Design of deck slabs
10.4.1 General—This section covers thin deck slabs
supported by beams for highway bridges, and deck slabs
which are the longitudinal load carrying members. A deck
slab may be designed by any procedure satisfying the condi-
tions of equilibrium and geometrical compatibility, provided
it is shown that the strength furnished is at least that required
by the criteria of Chapter 7, and that the design meets the
serviceability criteria of Chapter 8.
In the case of beam-supported thin deck slabs having
appropriate confinement in the plane of the slab, an allow-
ance may be made for the favorable effect of arching action
on slab strength.10-22—10-27 Where arching action cannot be
included, the following empirical methods may be used.
10.4.2 Empirical methods
10.4.2.1 Limitations—The method described here
applies to monolithic concrete deck slabs carrying highway
loads (AASHTO-1983). It should not be used where the
skew angle exceeds 20 deg (0.35 rad).
For greater skew, the slab design should be based on
model tests, a rational analysis based on principles of the
theory of plates, or other methods.10-20,10-28–10-30 The rein-
forcement at the obtuse corners of skew spans should receive
special consideration.10-31
10.4.2.2 One-way slab—The design of one-way
slabs10-20 should be based on the analysis of a strip of unit
width at right angles to the supports considered as a rectan-
gular beam. Bending moment per unit width of slab, due to
a standard truck load, should be calculated according to
empirical formulas given below, unless more exact methods
are used. In the following:
M = live load moment per foot width of slab, in ft-lb
(kNm/m)
en = clear span length of slab, in ft (m)
E = effective width of slab resisting a wheel or other
concentrated load, in ft (m)
343R-118 MANUAL OF CONCRETE PRACTICE
P = load on one rear wheel of truck, in lbs (N): 12,000
lbs for HS 15 loading (54 kN), 16,000 lbs for HS
20 loading (72 kN)
a. Main reinforcement perpendicular to traffic [spans 2.0
to 24 ft (1.7 to 7.3 m), inclusive]. The live load moment
for simple spans should be determined by the following
formula
M (lb ft/ft) = (en (ft) + 2) P(lb)/32
[M (kNm/m) = (en (m) + 0.61) P (kN)/9.7]
In slabs continuous over three or more supports, 80
percent of the previous calculated value should be used for
both positive and negative.
b. Main reinforcement parallel to traffic. Longitudinally
reinforced slabs should be designed for the appropriate
HS truck or lane loading, whichever causes larger
design moment. The effective width of slab resisting a
wheel load should be estimated to be E (ft) = 4 + 0.06
en (ft), [E (m) = 1.22 + 0.06 en (m)], but not to exceed
7.0 ft (2.1 m). The effective slab width resisting lane
load should be 2E.
For simple spans, the maximum live load moment per ft
width of slab, without impact, may be approximated by the
following formulas
HS 20 loading:
Spans up to and including 50 ft (15 m):
M (lb ft/ft) = 900 en (ft);
[M (kNm)/m) = 13.1 en (m)]
Spans 50 to 100 ft (15 to 30 m):
M (lb ft/ft) = 1000 [1.3 en (ft) – 20];
{M (kNm/m) = 14.6 [1.3 en (m) – 6.1]}
HS 15 loading:
Use 75 percent of the values obtained from the formulas
for HS 20 loading. Moments in continuous spans should be
determined by a suitable analysis using the truck or appro-
priate lane loading.
Edge beams should be provided for all slabs having main
reinforcement parallel to traffic. The beam may consist of a
slab section additionally reinforced, a beam integral with and
deeper than the slab, or an integral reinforced section of slab
and curb. The edge beam should be designed to resist a live
load moment of 0.10 Pen for simple spans. For continuous
spans, 80 percent of the previous calculated value should be
used for both positive and negative moments, unless a
greater reduction results from a more exact analysis.
10.4.2.3 Two-way slab—Two-way slabs are supported
on all four sides by beams, girders, or walls and are rein-
forced in both directions. For rectangular slabs simply
supported on all four sides, the proportion of the load carried
by the short span of the slab may be estimated by the
following equations:10-32
For load uniformly distributed, p = e24 /(e14 + e24 );
For load concentrated at center, p = e23 /(e13 + e23 );
where
p = proportion of load carried by short spans
e1 = length of short span of slab
e2 = length of long span of slab
Where e2 exceeds one-and-one-half times e1, the entire
load should be assumed to be carried by the short span. For
a concentrated load, the effective slab width E for the load
carried in either direction should be determined in accor-
dance with Section 10.3.2(a) and (b).
Moments obtained should be used in designing the center
half of the short and long spans. The area of reinforcement in
the outer quarters of both short and long spans may be 50
percent of that required in the center. For other supporting
conditions at the edges, the formulas for p can be adjusted to
account for the restraining effects at the edges. At the ends of
the bridge and at intermediate points where the continuity of
the slab is broken, the edges should be supported by
diaphragms or other suitable means.
10.4.2.4 Ribbed slabs—A two-way system consisting of
a slab with closely spaced ribs of similar size, equally spaced
in two directions, may be analyzed by the empirical methods
as an equivalent two-way slab. For two-way systems in
which the ribs are not equally spaced or are of different size,
an elastic analysis may be used in which the system is treated
as an equivalent orthotropic plate or a grid.10-19 A model
analysis may also be used.
10.4.2.5 Cantilever slabs
a. Truck loads. Cantilever slabs may be designed to
support truck wheel loads, ignoring the effect of edge
support along the end of the cantilever.
1. Main reinforcement perpendicular to traffic. The
effective slab width in ft (m), perpendicular to
traffic, for each wheel load should be E = 0.8 x1 +
3.75 (0.8 x1 + 1.14), where x1 = distance in ft (m)
from load to point of support.
2. Main reinforcement parallel to traffic. The effec-
tive slab width, parallel to traffic, for each wheel
load should be E = 0.35 x1 + 3.2 (0.35 x1 + 0.98),
but not to exceed 7.0 ft (2.1 m).
b. Railing loads. The effective width of slab resisting
railing post loadings should be estimated as E = 0.8 x2
+ 3.75 (0.8 x2 + 1.14) if no parapet is used, and E = 0.8
x2 + 5.0 (0.8 x2 + 1.52) if a parapet is used. In the
previous expressions, x2 is the distance in ft (m) from
the center of the post to the point under investigation.
10.5—Distribution of loads to beams
Analysis based on elastic theory is recommended to find
the transverse distribution of the bending moment. For the
analysis, the structure may be idealized in one of the
following ways:
a. A system of interconnected beams forming a grid.
b. An orthotropic plate.
 BRIDGE ANALYSIS AND DESIGN
c. An assemblage of thin plate elements or thin plate
elements and beams.
Several methods of analysis are available10-19 which can
be applied with the use of a computer. In addition to the
moments in the direction of the span, the computer-aided
analysis can give moments in the transverse members. A
theoretical analysis is particularly recommended for bridges
which have large skew or sharp curvature.
10.5.1 T-beam or precast I-girder and box girder
bridges—In lieu of an analysis based on elastic theory for the
distribution of live loads to longitudinal beams, the
following empirical method authorized by AASHTO (1983)
may be used for T-beam or precast I-girder bridges and for
box-girder bridges. The distribution of shear should be deter-
mined by the method prescribed for moment.
10.5.1.1 Interior beams—The live load bending
moment for each interior longitudinal beam should be deter-
mined by applying to the beam the distributing fraction (DF)
of the wheel load (both front and rear) as defined in Table
10.5.1.1.
10.5.1.2 Exterior beams
a. T-beam or precast I-girder bridges. The dead load to be
considered as supported by the outside roadway beam
is the weight of the beam plus that portion of the floor
slab carried by the stringer or beam. Curbs, railings,
and wearing surface, if placed after the slab has cured,
may be considered equally distributed to all roadway
stringers or beams.
When the outside roadway beam supports the side-
walk live load, as well as traffic live load and impact,
the allowable stress in the beam may be increased 25
percent for the combination of dead load, sidewalk live
load, traffic live load, and impact, providing the beam
has no less carrying capacity than would be required if
there were no sidewalks. To permit future widening, an
exterior beam should never have less carrying capacity
than an interior beam.
b. Box girder bridges. The dead load considered as
supported by the exterior girder should be determined
in the same manner as given previously in (a).
The wheel load distribution to the exterior girder
should be we/7 (we/2.13). The width we to be used in
determining the wheel load distribution to the exterior
girder, is the top slab width measured in ft (m) from the
midpoint between girders to the outside edge of the
slab. The cantilever dimension of any slab extending
Table 10.5.1.1—Distribution factor for bending moments in interior main beams
Kind of floor
T-beam and precast I-girder bridges S/6.5 (S/1.98). If S exceeds 6 ft (1.8 m), use
Box-girder bridges S/8.0 (S/2.44). If S exceeds 12 ft (3.7 m), use
Spread box-beam bridges
Multibeam precast concrete beams (adjacent)
* In this case, the load on each beam will be the reaction of the wheel loads assuming the flooring between the beams to act as a simple beam.
S = average beam spacing in ft (m).
343R-119
beyond the centerline of the exterior girder should not
exceed S/2, where S is the girder spacing.
10.5.1.3 Total capacity of longitudinal beams—The
combined design load capacity of all the beams and stringers
in a span should not be less than required to support the total
load in the span.
10.5.1.4 Bending moments for T- and I-girder bridges
with cross girders—The following equations are applicable
for T- and I-girder rectangular (non-skew) bridges, simply-
supported or continuous.10-33 Each span should be provided
with at least one cross girder at the center. The depth of the
cross girder should be greater than or equal to three-fourths
the depth of the main girder.
The wheel loads, both front and rear, are to be multiplied
by a distribution factor DF, and the resulting loads applied to
a single beam in a position which gives maximum bending
moment. The distribution factor for an interior or exterior
main girder is given by
DF = 2 (ne/nb) + k S/e
where
ne = number of design traffic lanes
nb = number of girders, 3 ≤ nb ≤ 10
S = girder spacing in ft (m), where 6 ≤ S ≤ 16, (2 ≤ S ≤ 5)
e = span in ft (m) for a simply-supported bridge
e = distance between points of inflection under uniform
load for a continuous bridge
The dimensionless coefficient k is given by one of the
following equations
For nb = 3 to 5,
k = 0.08 wc – ne (0.10 ne – 0.26) – 0.20 nb – 0.12
For nb = 6 or 7,
k = 0.09 wc – ne (0.10 ne – 0.26) – 0.20 nb – 0.12
For nb = 8 to 10,
k = 0.10 wc – ne (0.10 ne – 0.26) – 0.20 nb – 0.12
where
wc = roadway width between curbs in ft (m), 20 ≤ wc ≤
78, (6 ≤ wc ≤ 24).
Distributing factor (DF)
Bridge designed for one traffic lane Bridge designed for two or more traffic lanes
*
S/6.0 (S/1.83). If S exceeds 10 ft (3.0 m), use *
*
S/7.0 (S/2.13). If S exceeds 16 ft (4.9 m), use *
See Section 10.5.2
See Section 10.5.3
343R-120 MANUAL OF CONCRETE PRACTICE
When the SI system of units is used, wc will be in meters
and the coefficients of wc in the equations of k will be 0.26,
0.30 and 0.33, respectively.
10.5.2 Spread box-beam bridges—For spread box-beam
superstructures, the lateral distribution of live load for
bending moment can be determined as follows:10-34
a. Interior beams. The live load bending moment in each
interior beam can be determined by applying to the
beam the fraction (DF) of the wheel load (both front
and rear) determined by the following equation
DF = 2ne/nb + ke S/e
where
ne = number of design traffic lanes
nb = number of beams, 4 ≤ nb ≤ 10
S = beam spacing in ft (m), where 6.75 ≤ S ≤ 11.00 (2 ≤
S ≤ 3.35)
e = span-length in ft (m) (for continuous structures use
length between points of inflection under uniform
load);
ke = 0.07 wc – ne (0.1 ne – 0.26) – 0.20 nb – 0.12
wc = roadway width between curbs in ft (m), 32 ≤ wc ≤
66 (9.7 ≤ wc ≤ 20)
When the SI system of units is used, wc will be in meters
and the coefficient of wc in the equation for ke will be 0.23.
b. Exterior beams. The live load bending moment in the
exterior beams should be determined by applying to the
beams the reaction of the wheel loads obtained by
assuming the flooring to act as a simple span between
beams, but the reaction should not be less than 2ne/nb.
10.5.3 Multi-beam precast concrete bridge—A multi-
beam bridge is constructed with precast nonprestressed or
prestressed concrete beams which are placed in contact side
by side on the supports. The interaction between the beams
is developed by continuous longitudinal shear keys and
lateral bolts which may, or may not, be prestressed.
In calculating bending moments in multibeam precast
concrete bridges, conventional or prestressed, no longitu-
dinal distribution of wheel load should be assumed. The
lateral distribution should be determined by the following:
The live load bending moment for each section should be
determined by applying to the beam the fraction of a wheel load
(both front and rear), determined by the following relations
DF = [(12ne + 9)/nb])/(5 + ne/10) + [(3 – 2ne/7) (1 – C/3)2]
or when C is 3 or greater, use
DF = [(12ne + 9)/nb]/(5 + ne/10)
where
ne = number of design traffic lanes
nb = number of beams
C = K(wc/e), a stiffness parameter
wc = roadway width between curbs in ft (m)
e = span length in ft (m) (for continuous structures, use
length between points of inflection under uniform
load).
Values of K to be used in C = K(wc/e):
Beam type and deck material K
Nonvoided rectangular beams 0.7
Rectangular beams with circular voids 0.8
Box section beams 1.0
Channel beams 2.2
10.5.4 Transverse floor beams—In calculating bending
moments in floor beams, no transverse distribution of the
wheel loads should be assumed. If longitudinal beams are
omitted and the slab is supported directly on floor beams, the
floor beams should be designed for live loads determined in
accordance with the following formula: fraction of wheel
load to each beam = S/6 (S/1.83), where S = spacing of beams
in ft (m). If S exceeds 6 ft (2 m), the load on the beam should
be the reaction of the wheel loads, assuming the flooring
between beams is acting as a simple beam.
10.5.5 Position of loads for shear—In calculating end
shears and end reactions in transverse floor beams and longi-
tudinal beams, no longitudinal distribution of the wheel load
should be assumed for the wheel or axle load adjacent to the
end at which the stress is being determined.
Lateral distribution of the wheel load should be produced
by assuming the flooring to act as a simple span between
stringers or beams. For loads in other positions on the span,
the distribution for shear is determined by the method
prescribed for moment.
10.6—Skew bridges
10.6.1 General—Empirical or simplified methods of anal-
ysis mentioned in Section 10.3 can be used for skew bridges,
provided adjustments for skew are made. The analysis may
be performed by one of the methods mentioned in Section
10.3.2b and in Section 10.3.3. Design tables based on
such methods may be employed.10-28,10-34
10.6.2 Bending moments—The bending moment in the
longitudinal direction in a skew bridge is generally smaller
than the bending moment in a rectilinear bridge of the same
span.10-35 However, torsion moments exist about the longitu-
dinal axis in skew bridges, increasing shear forces in the
vicinity of the obtuse corners.
10.6.3 Reactions—The reactions and shear due to dead load
in a skew bridge are higher at the obtuse angle compared to the
values at an acute angle. The nonuniformity in distribution of
reaction or shear increases with the increase in skew angle.
The risk of the reaction becoming very small or negative at an
acute angle should be avoided. This can be achieved in post-
tensioned bridges by appropriate choice of the prestressing
forces and the tendon profiles. The higher reactions at the
obtuse corners should be accounted for in the design of bear-
ings and the supporting elements.
The California Department of Transportation has devel-
oped the chart shown in Fig. 10.6.3 to adjust the shears and
reactions in skew bridges. The factors shown are applied to
the shears and reactions calculated for a rectilinear bridge
having the same span. For shear design the factor is assumed
 BRIDGE ANALYSIS AND DESIGN
to vary linearly from the maximum shown at the support to
unity at midspan. These factors are applied to all spans of a
skew bridge regardless of end condition. Curved bridges
having large skews over 45 deg should be analyzed by an
exact method.
End diaphragms in skew bridges, either box or multi-
girder, increase the difference between reactions and
multiply the risk of having negative reactions. A diaphragm
at midspan has little beneficial effect in transmitting eccen-
tric live load from one girder or web to another. Additional
diaphragms have negligible effects. In some cases, the
diaphragms in box girder skew bridges may have a harmful
effect. Thus, in view of the extra time and the cost they
require, the diaphragms may be dispensed with.10-36
RECOMMENDED REFERENCES
The documents of the various standards-producing organi-
zations referred to in this report are listed below with their
serial designation, including year of adoption or revision.
The documents listed were the latest effort at the time this
report was written. Since some of these documents are
revised frequently, generally in minor detail only, the user of
this report should check directly with the sponsoring group
if it is desired to refer to the latest revision.
American Association of State Highway and Transportation
Officials
Standard Specifications for Highway Bridges, Thirteenth
Edition, 1983.
American Concrete Institute
SP-24 Models for Concrete Structures
SP-35 Analysis of Structural Systems for Torsion
CITED REFERENCES
10-1. Bassi, K. G.; Lin, W.; and Richardson, B. S.,
“Continuous, Post-Tensioned, Torsionally Stiff Concrete
Bridges,” International Symposium on Concrete Bridge
Design, Toronto, Apr. 1967.
10-2. Campbell, T. I.; Francis, L. N.; and Richardson,
B.S., “A Long-Curved Post-Tensioned Bridge Without
Expansion Joints,” Canadian Journal of Civil Engineering,
V. 2, No. 3, 1975, pp. 262-269.
10-3. Csagoly, P., and Holowka, M., “Cracking of Voided
Post-Tensioned Concrete Bridge Decks,” Ministry of Trans-
portation and Communications, Ontario, RR 193, 1974.
10-4. Elbadry, M., and Ghali, A., “Nonlinear Temperature
Distribution and Its Effects on Bridges,” International Asso-
ciation for Bridges and Structural Engineering, Periodical
3/1983, Proceedings, P-66/83, pp. 169-191.
10-5. Ghali, A., and Favre, R., Concrete Structures: Stresses
and Deformations, Chapman and Hall, London, 1986.
10-6. Tadros, M. K.; Ghali, A.; and Dilger, W. H., “Long-
Term Stresses and Deformations of Segmental Bridges,”
Journal of the Prestressed Concrete Institute, V. 24, No. 4,
July-Aug. 1979, pp. 66-89.
10-7. Scordelis, A. C.; Davis, R. E.; and Lo, K. S., “Load
Distribution in Concrete Box Girder Bridges,” International
343R-121
Fig. 10.6.3—Shear factors for skew bridges
Symposium on Concrete Bridge Design, Toronto, Apr.
1967.
10-8. Scordelis, A. C., and Meyer, C., “Wheel Load Distri-
bution in Concrete Box Girder Bridges,” College of Engi-
neering, University of California, Berkeley, January 1969.
10-9. Sisodiya, R. G., and Ghali, A., “Analysis of Box
Girder Bridges of Arbitrary Shape,” Publications, Interna-
tional Association of Bridges and Structural Engineering, V.
33-I, 1973, pp. 203-218.
10-10. “Precast Segmental Box Girder Bridge Manual,”
Prestressed Concrete Institute and Post-Tensioning Institute,
1978.
10-11. Libby, J. R., “Segmental Box Girder Bridge Super-
structure Design,” ACI JOURNAL, Proceedings V. 73, No.
23, May 1976, pp. 279-290.
10-12. Ballinger, C. A.; Podolmy, W.; and Abrahams, M.
J., “A Report on the Design and Construction of Segmental
Prestressed Concrete Bridges in Western Europe—1977,”
FHWA, Washington, D.C.
10-13. “Recommended Practice for Segmented Construc-
tion in Prestressed Concrete,” PCI Journal, Mar.-Apr. 1975.
10-14. Mathivat, J., The Cantilever Construction of
Prestressed Concrete Bridges, John Wiley and Sons, New
York, 1983.
10-15. Podolny, Jr., Walter, “Cable-Stayed Bridges of
Prestressed Concrete,” PCI Journal, V. 18, No. 1, Jan.-Feb.
1973, pp. 68-79; Closure—PCI Journal, V. 18, No. 5, Sept.-
Oct. 1973, pp. 115-116.
10-16. Van De Pitte, D., “Prestressed Concrete Suspen-
sion Bridges,” Proceedings on World Conference of
Prestressed Concrete, 1967.
10-17. Dudra, J., “Design and Construction of Hudson
Hope Bridge,” Journal of the Prestressed Concrete Institute,
Apr. 1966.
10-18. Tung, David H. H., “Torsional Analysis of Single
343R-122 MANUAL OF CONCRETE PRACTICE
Thin-Walled Trapezoidal Concrete Box Girder Bridges,”
International Symposium on Concrete Bridge Design,
Toronto, Apr. 1967.
10-19. Jaeger, L. G., and Bakht, B., “The Grillage
Analogy in Bridge Analysis,” Canadian Journal of Civil
Engineering, V. 9, No. 2, June 1982, pp. 224-235.
10-20. Brown, T. G., and Ghali, A., “Semi-Analytic Solu-
tion of Skew Plates in Bending,” Proceedings, Institution of
Civil Engineers, London, Mar. 1974, pp. 19-22.
10-21. Mehmel, A., and Weise, H., “Model Investigation
of Skew Slabs on Elastically Yielding Point Supports,”
Cement and Concrete Association, London, 1965.
10-22. Ontario Highway Bridge Design Code, V. II,
Supplement S7, Ministry of Transportation and Communi-
cations, Ontario, 1979.
10-23. Hewitt, B. E., and Batchelor, B. deV., “Punching
Shear Strength of Restrained Slabs,” Journal of the Struc-
tural Division, ASCE, V. 101, ST9, Sept. 1975, pp. 1937-53.
10-24. Dorton, R. A.; Holowka, M.; and King, J. P. C., “The
Conestogo River Bridge—Design and Testing,” Canadian
Journal of Civil Engineering, V. 4, No. 1, 1977, pp. 18-39.
10-25. Batchelor, B. deV.; Hewitt, B. E.; Csagoly, P.; and
Holowka, M., “Investigation of the Ultimate Strength of
Deck Slabs of Composite Steel/Concrete Bridges,” Trans-
portation Research Record 664, Bridge Engineering, V. 1,
Transportation Research Board, Sept. 1978.
10-26. Ontario Highway Bridge Design Code, Second
Edition, Ministry of Transportation and Communications,
Ontario, 1983.
10-27. Ontario Highway Bridge Design Code, Commen-
tary, Ministry of Transportation and Communications,
Ontario, 1983.
10-28. Rusch, H., and Hergenroder, A., “Influence
Surfaces for Moments in Skew Slabs,” Cement and Concrete
Association, London.
10-29. “Analysis of Continuous Skewed Slab Bridge
Structures,” Ohio Department of Highways, 1961.
10-30. Schleicher, C., and Wegener, Borwin, “Continuous
Skew Slabs,” Veb Verlag Lur Bauwesen, Second Edition
Berlin.
10-31. Kennedy, J. B., and Tamberg, K. G., “Problems of
Skew in Concrete Bridge Design,” Report No. 44 144,
Department of Highways, Ontario.
10-32. Culham, G. A., and Ghali, A., “Distribution of
Wheel Loads on Bridge Girders,” Canadian Journal of Civil
Engineering, V. 4, No. 1, 1977, pp. 57-65.
10-33. Motarjemi, D., and Van Horn, D. A., “Theoretical
Analysis of Load Distribution in Prestressed Concrete Box-
Beam Bridges,” Fritz Laboratory Report No. 315.9, Lehigh
University, Oct. 1969.
10-34. Ghali, A., “Design of Simply-Supported Skew
Concrete Girder Bridges,” Second International Symposium
on Concrete Bridge Design, SP-26, American Concrete
Institute, Farmington Hills, Mich., 1969, pp. 206-234.
10-35. Ghali, A., “Analysis of Continuous Skew Concrete
Girder Bridges,” Concrete Bridge Design, Proceedings of
Symposium, SP-23, American Concrete Institute, Farmington
Hills, Mich., pp. 137-169.
10-36. Sisodiya, R. G., and Ghali, A., “Diaphragms in
Single and Double Cell Box Girder Bridges with Varying
Angle of Skew,” ACI JOURNAL, Proceedings V. 69, No. 7,
July 1972, pp. 415-419.
 BRIDGE ANALYSIS AND DESIGN 343R-123

Overcrossing in California with curved alignment and curved soffit (note flared integrated pier and unobtrusive abutments).

CHAPTER 11—SUBSTRUCTURE SYSTEMS AND ELEMENTS

11.1—Introduction including 400 ft (122 m), with maximum reactions of about

The quality of a bridge project, its visual impact, and
economic viability is not determined solely by the type of
superstructure chosen. Of almost equal importance is the
proper selection of the substructure system and the details of
the elements for that system. This chapter is intended to be a
guide to making the selection as the designer proceeds
through the bridge design process. Guidance is supplied for
abutments, piers, foundations, and pier protection.
Connections of substructure units to superstructure
systems can be accomplished by several different methods.
The subject of bearings and other types of connections
typical to concrete structures is addressed in this chapter.
11.2—Bearings
11.2.1 Description—A wide variety of bearings is avail-
able for bridge structures. The main purpose of any bearing
is to transfer the superstructure load, both vertical and hori-
zontal. In addition, the bearing should accommodate transla-
tional and/or rotational movements. In some cases the
bearing will transmit eccentric loads.
11.2.2 Types and design criteria
11.2.2.1 Elastomeric bearings (Fig. 11.2.2.1)—Elasto-
meric bearing pads have been used in the U.S. since about
1960. They are popular in the short span range up to and
400 kips (1800 kN). They are economical and have proven
to be excellent in service.
Two main types of elastomeric bearings are used; plain
pads, consisting of elastomer only, and reinforced pads,
consisting of layers of elastomer restrained at their interfaces
with integrally bonded steel or fabric reinforcement. The
elastomer material may be polyisoprene (natural rubber) or
polychloroprene (neoprene).
Tapered elastomeric bearings should not be used. Where
necessary, tapered wedge steel plates should be provided. In
some cases the wedge should be tapered transversely, as well
as longitudinally. In locations where it is expected that the
bearings will be subjected to temperatures below -25 F (-32
C) for several days, consideration should be given to speci-
fying natural rubber and requiring special testing by the
manufacturer for the temperature range expected.
Because of their compressibility, elastomeric bearings can
readily accommodate rotational movements anticipated in a
normal type structure. In relatively short bridges the longitu-
dinal movement caused by temperature changes is accom-
modated by the shear deflection of the bearing, as shown in
Fig. 11.2.2.1(a). When the expected longitudinal movement
exceeds the allowable shear deflection, sliding plates can be
installed, as shown in Fig. 11.2.2.1(b).
343R-124 MANUAL OF CONCRETE PRACTICE
Fig. 11.2.2.1—Elastomeric bearings.
If some combination of loads is possible which would
cause a shear force greater than one-fifth of the simulta-
neously occurring compression force, or if the dead load
stress in the bearings is less than 200 psi (1.38 MPa), the
bearing should be secured against horizontal movement.
When the bearing is anchored at both the top and bottom
surfaces, the anchorage should be such that no tension is
possible in the vertical direction. One example is shown in
Fig. 11.2.2.1(c). If welding is used as a means of attachment,
care should be taken to limit the temperature in the plate
adjacent to the elastomer to 400 F (204 C).
The Transportation Research Board (TRB) of the National
Research Council has funded research into the design and
construction of elastomeric bearings. The first report
(NCHRP No. 248) was published in 1982. The project is
being continued and state-of-the-art reports will be issued.
11.2.2.2 Sliding bearings [Fig. 11.2.2.1(c)]—When the
expected expansion or contraction of the bridge superstruc-
ture due to temperature changes is so large that the corre-
sponding elastic deflection of the pier top would produce too
great a horizontal force, sliding bearings should be used.
This is particularly true when the piers are short and heavy.
Sliding bearings are also used at abutments to accommodate
movements due to earth pressures.
Sliding bearings may be fabricated with plain or rein-
forced elastomeric pads bonded to a polytetrafluorethylene
Fig. 11.2.2.3—“High” load bearings.
(PTFE) surface. The underside of the sole plate should have
a stainless steel plate to provide the proper sliding surface.
11.2.2.3 “High” load bearings—In straight bridges with
right angle piers, the rotational movement that should be
accommodated occurs about one easily predicted axis. In
curved bridges, or straight bridges with large skews, the rota-
tional movement occurs about more than one axis and is not
readily predictable. To resolve this problem, “pot” bearings
having the ability to accommodate rotational movement about
any axis that has been developed. Fig. 11.2.2.3 illustrates
graphically the main elements of this type of bearing. The
term “pot” is derived from the shape of the bearing base plate.
As shown, the bearing can provide free translation in any
direction, as well as rotational freedom [Fig. 11.2.2.3(a)], or it
may be restrained against translation in one direction, [Fig.
11.2.2.3(b)], or all directions [Fig. 11.2.2.3(c)].
The pot bearing also permits the use of high allowable
bearing stresses since the confined elastomer acts like a
hydraulic fluid. These bearings are manufactured by a
number of companies, and the bridge designer should main-
tain a file of the latest information furnished.
11.2.2.4 Steel bearings—In the past, all bridge bearings
consisted of structural steel plates and shapes. Many of these
types of bearings are still being used. Fig. 11.2.2.4(a) shows
a typical roller bearing suitable for use on short span bridges.
Because the bearing capacity is directly related to the diam-
eter of the roller, this type has a very limited capacity of about
50 kips (220 kN). Fig. 11.2.2.4(b) shows a segmental roller
 BRIDGE ANALYSIS AND DESIGN
Fig. 11.2.2.4—Steel bearings.
bearing used on cast-in-place concrete girder bridges. Since
the diameter of the bearing surface is much greater than a
roller bearing of the same height, and the weight is much less
than a solid cylinder, the bearing can be economically used to
carry significant loads of up to 250 kips (1100 kN).
For larger loads, rocker-type bearings as shown in Fig.
11.2.2.4(c) can be designed. The diameter of the bearing
surface can be tailored to fit the need. The material can be
either steel castings or structural steel weldments.
The bearings shown in Fig. 11.2.2.4 are expansion bearings
providing longitudinal movement. Transverse movement is
restricted. In the case of the segmental roller and rocker, the
permissible longitudinal movement is limited and care should be
taken in setting the bearings to adjust the inclination of the bearing
to suit the actual temperature of the bridge at the time of setting.
Fixed bearings for bridges that have segmental roller or
rocker-type expansion bearings are generally the same type,
343R-125
but with the rocker or roller has a flat bottom and is welded
to the masonry plate.
11.2.2.5 Bearings in seismic zones—At locations
requiring the bridge to be designed for seismic forces and
movements, bearings are the most critical component of the
bridge. Discussion of the forces and movements is given in
Chapter 5, and the designer should give special consider-
ation to the latest state-of-the-art criteria. Guidance is given
in ATC 611-1 and in AASHTO publication GSDB. Without
proper modification, all of the bearing types previously
discussed are unsuitable for use in seismic areas.
11.2.2.6 General criteria—The allowable bearing pres-
sure on concrete should be 0.30fc′ (see AASHTO for further
requirements). The steel masonry plate should be designed
for this allowable bearing stress using an allowable flexural
stress of 0.55fy.
343R-126 MANUAL OF CONCRETE PRACTICE
Fig. 11.3.3—Net bearing pressure.
One or more fixed bearings should be provided in each
bridge structure. A fixed bearing is designed to prevent any
horizontal movement between the superstructure and
substructure.
Expansion bearings should be designed to accommodate the
contributory movement due to elastic shortening, creep,
shrinkage, and temperature multiplied by 125 percent. Where
lateral restraint is desired in the bearings, the bearing restraints
should be designed for 15 percent of the axial dead load.
11.3—Foundations
11.3.1 General—The function of a foundation is to
transfer the loads from the bridge elements above ground to
the soil or rock without objectionable vertical or lateral
movement. Selection and design of the proper foundation is
dependent upon the design structural loading conditions,
bridge pier geometry, and surface and subsurface conditions
at the site, plus interpretation of field data and laboratory
tests combined with engineering judgment.
The foundation should be located as high as possible, since
costs increase significantly with depth. Even though costs do
increase significantly with depth, the foundation elevation
should be deep enough to provide long term bearing support
to prevent undue total and differential settlement to resist
lateral forces and to avoid problems from ground movement
(e.g. frost heave), scour, and/or future deepening of a
channel area. Depending on site conditions, this may most
economically be accomplished by open excavation, driving
piles, or drilling piers.
11.3.2 Investigation procedures—An excellent guide to
the type of information that should be obtained is provided
in the AASHTO Foundation Investigation Manual (1978).
Additional information on soils engineering and foundation
design may be found in References 11-2 through 11-16 and
ASTM Special Technical Publications No. 444 and 670. It is
recommended that the structural engineer work closely with
a geotechnical engineer experienced in the requirements of
foundations that are unique to bridges. The following is a
partial list of the information that should be obtained (Refer-
ences 11-17 through 11-18 and ASTM Special Reports No.
399, 412 and 479):
a. Distribution, classification and physical properties of
the soil and underlying rock.
b. Determination of the allowable bearing capacity of the
selected foundation.
c. Determination of design lateral earth pressure for the
abutment walls and wingwalls.
d. Prediction of settlement of the foundation soils.
e. Determination of high and low water levels, velocity of
stream, and depth of scour. Prediction of the future
lowering of the stream bed and future scour depth
during the life of the bridge for stream crossings.
f. Determination of ground water level and depth of frost
line or depth of zone of significant volume change in
expansive clay subsoils.
g. Location, depth, type, and possible influence of the
foundations of adjacent existing or future structures or
underground utilities.
h. Stability of abutment slopes.
11.3.3 Spread footings—Spread footings are classified as
shallow when the depth of foundation below the surrounding
final ground line is less than the least width of the footing or
deep when depth of foundation is greater than the width of
the footing. Shallow footings are generally the most
economical and should be considered unless ruled out by
other dominant factors.
Shallow spread foundations should be proportioned so that
the maximum net contact pressure under the combined
effects of dead, live, and transient loads does not exceed the
allowable net bearing pressure determined by the geotech-
nical engineer. Net bearing pressure is defined in Fig. 11.3.3.
Foundation movement should be estimated by the geotech-
nical engineer for the appropriate loading conditions. The
expected foundation movement should be within tolerable
limits for the bridge structure. The allowable net bearing
pressure may be limited by the movement requirements,
either vertical or lateral.
Many superstructure types are sensitive to the differential
settlement of piers. In general, simple spans are not affected,
and long continuous spans are less sensitive than short spans.
Especially sensitive are arches or short continuous spans
where adjacent piers are founded on subsoils that vary greatly.
Settlement of footings, particularly abutment footings,
may be influenced not only by the load from the bridge
superstructure, but by the fill supporting the approach
roadway. Settlement can also be caused by vibratory load-
ings, such as railroad loadings on shallow footings, which
bear on loose to medium dense granular soils. An estimate
should be made of the differential settlement that the
proposed superstructure can accommodate. This estimate
should be used by the geotechnical engineer in determining
the allowable soil pressure or pile capacity.
Spread footings for bridges located in seismic areas should
be designed using the criteria given in ATC-6, “Seismic
Design Guidelines for Highway Bridges.11-1” Additional
information can be found in ASTM Special Technical Publi-
cation No. 450, 1969.
11.3.4 Drilled piers
11.3.4.1 General—Drilled piers are, in effect, deep
spread footings. The development of special equipment for
their installation has made drilled pier foundations time
 BRIDGE ANALYSIS AND DESIGN
saving as well as economical. It is expected that their use in
bridges will become more widespread. Although primarily
used to transfer the bridge loads to a deep strong layer such
as hardpan or rock by end bearing, drilled piers can function
by transferring the load to the soil through friction. This
ability is seldom considered in the design, since its neglect
results in a more conservative design, but when the possi-
bility for negative skin friction exists, as in drilled piers
supporting abutments, the additional load due to negative
skin friction should be included in sizing the bearing area.
ACI 336.3R describes in detail the design and construction
procedures for drilled piers.
The cost of drilled piers is directly related to the amount of
material excavated and concrete placed, so it is advanta-
geous to minimize the size of the shaft. Most drilling equip-
ment can construct a bell having a diameter equal to three
times the shaft diameter. Bells can be enlarged by hand, but
this is expensive. The condition of the soil at the bottom of
the pier should be inspected by a qualified soils engineer.
The minimum size of the shaft to permit this is 3 ft (0.9 m).
The shaft diameter should be selected with consideration
of moment capacity, the placing and spacing of reinforcing
steel, the placement of concrete, and the necessary cover
over the steel.
11.3.4.2 Construction consideration—Since drilled
piers are usually constructed from the ground level prior to
excavation for the foundation, allowance should be made for
a reasonable tolerance in location. This is usually specified
as 4 percent of the diameter of the shaft. To accommodate
this tolerance, dowels for the column extension above the
drilled pier should not be set directly in the subpier shaft, but
in a cap constructed after the excavation has been completed.
This permits accurate location of the column dowels.
Drilled piers are often subjected to horizontal loads, some
anticipated in the design and some caused by construction
operations, unforeseen soil conditions or future construction,
such as trenches. For this reason, a minimum amount of rein-
forcing, usually 0.5 percent of the gross area of the shaft,
should be provided for the full height or nearly the full height
of the shaft.
11.3.4.3 Classes of subpiers—Drilled piers can be classi-
fied as belled piers, which are enlarged at the base to increase
the bearing area, and rock-socketed piers, which have sockets
drilled into solid rock to increase the end bearing capacity. The
belled piers are generally more economical and are always
used unless the spacing of the piers (bridge columns) causes
the bells to overlap. Belling can be done only if there is an
adequate layer of cohesive material just above the proposed
bottom of the drilled pier and there is no bearing seam in that
layer. New techniques in soil testing, such as in-situ pressure
meters that measure the in-place strength of the bearing layer,
have permitted the use of larger allowable bearing pressures.
Allowable pressures as high as 30 kips/ft2 (1.45 MPa) have
been used to design pier bells founded on hard silt layers.
11.3.5 Piles
11.3.5.1 General—Timber piles have been used for
centuries, and excavations at old building sites reveal that
343R-127
they are in good condition when located below the perma-
nent ground water table. Iron or steel piles have been used
since the middle of the nineteenth century, and since early in
the twentieth century, special shapes called H piles and steel
pipe piles have become standard. Reinforced concrete piles
were developed at the turn of the century, and numerous
shapes and methods of installation have been invented and
patented. Prestressed concrete piles have been developed
recently and have been used extensively in structures where
large quantities of piles have been required. An excellent
description of the history of piles and pile driving is found in
“Pile Foundations.”11-19
While the primary function of a pile is to transmit the load
of the structure to a suitable soil stratum below the surface of
the ground without excavation, it also has other properties
which may or may not be a factor in the design, such as:
a. It is capable of anchoring structures against uplift or
overturning forces.
b. It may improve the properties of the soil through which
it is driven by consolidation.
c. It can resist lateral loads by bending as well as by direct
action through battered piles.
d. It can function as an unsupported column, where
support furnished by the upper soil layers is removed
by scour or an adjacent excavation.
11.3.5.2 Classes of piles—A pile can transmit the load
into the adjacent soil by skin friction or end bearing, and is
generally referred to as a friction or bearing pile. Actually,
there is usually friction on any pile, unless specific means are
taken to eliminate it, and there is likewise some end bearing.
However, it is common practice to design a pile as either
transmitting all its load through friction, neglecting end
bearing, or as solely end bearing and neglecting friction. For
large piles, both friction and end bearing capacities can be
used. Piles have also been used solely to compact the upper
strata of soil, but this use for bridge structures is rare.
11.3.5.3 Pile design—The design of piles for footings
and bents requires considerable experience and the best of
good engineering judgment. It is the designer's responsibility
to select types of piles which meet the field requirements
and, in addition, fulfill the requirements for economy, avail-
ability, and practicability for the particular conditions
prevailing at the site.11-19—11-30 Details for design of
concrete piles are given in ACI 543R.
For pile foundations in seismic areas, consideration should
be given to the fact that batter piles are much stiffer than
vertical piles under seismic loading. It may be desirable to
use only the more flexible vertical piles. This is discussed in
AASHTO’s “Guide Specifications for Seismic Design of
Highway Bridges.” Numerous failures have occurred during
earthquakes in concrete piles just below the pile cap. This is
probably due to the abrupt change of stiffness at that location
and the lack of confining lateral reinforcement in the pile.
The designer should consider these critical areas.
In designing and detailing the footing or cap for the piles,
consideration should be given to providing a reasonable
amount of tolerance in the location of the pile. It is reason-
able to expect the center of the pile group to be within 3 in.
343R-128 MANUAL OF CONCRETE PRACTICE
(75 mm) of the design location, but the location of each indi-
vidual pile need not be so restricted. A reasonable design
should assume that the individual piles could be out of loca-
tion by at least 6 in. (150 mm).
11.3.5.4 Design criteria—The design loads for piles
should not be greater than the least of the following values:
a. Capacity of the pile as a structural member.
b. Capacity of the pile to transfer its load to the ground.
c. Capacity of the ground to support the load transmitted
to it by the pile or piles.
Required penetrations are initially estimated on the basis
of soil borings and then substantiated by test piles, which are
piles driven by the contractor to ascertain the proper length
before ordering the remainder of the piles. These piles are
not used for load testing. The test piles are generally
furnished 10 ft (3 m) longer than the estimated length and
driven to refusal or to a capacity 50 percent greater than the
required design capacity. The capacity may be determined
by load tests or by dynamic analyses, such as the wave equa-
tion analysis, or by both. Other dynamic formulae may be
used with judgment, especially if correlated with a load test
of the wave equation. From the results of the test pile, the
engineer determines the proper length and capacity of the
permanent piles. In addition, consideration also should be
given to the following:
a. The difference between the supporting capacity of a
single friction pile, as against a group of friction piles.
b. The effect of driving piles and the pile loads on adja-
cent structures.
c. The possibility of scour and its effects.
d. The possibility of negative skin friction (“downdrag”).
The recommended values of allowable loads and stresses
for an acceptable design of concrete piles are listed in ACI
543R.
11.3.5.5 Pile load tests—Pile load tests may be recom-
mended for any of the following conditions:
a. At locations or in types of materials where the ordinary
methods of determining safe pile loads are not consid-
ered reliable.
b. For cast-in-drilled-hole piles in unproven soil forma-
tions.
c. Where it is desired to demonstrate that piles may safely
be loaded beyond the indicated safe loads obtained by
the application of standard pile driving formulae.
d. For very large or heavily loaded piles or for unusual or
innovative installations.
e. Critical uplift forces.
In general, the loading test should be carried out following
the procedures described in ACI 543R or ASTM D 1143.
11.3.5.6 Anchorage for uplift—The possibility of uplift
should be investigated.
Uplift capacity of a pile can best be determined by a tensile
pile field test. Often this can be done in conjunction with
load bearing tests by using reaction piles of the type and
length to be installed for the foundation.
The uplift capacity can be estimated by considering such
factors as soil shear strength, pile spacing, shape, length,
size, and weight, and weight of the soil surrounding the pile
or pile group.
Reinforced or prestressed concrete piles, which are
subjected to uplift, should have dowels or reinforcing bars
anchoring the piles to the footing. Strands of sufficient length
to develop the uplift force may be extended from prestressed
concrete piles into the footing. This is acceptable and common
practice. Timber piles should not be used when uplift is
expected, unless an adequate anchorage that will fix the pile to
the footing is provided, and provided the tapering of the timber
pile is taken into account in the design. Steel piles, when
subjected to uplift, should be provided with adequate
anchorage devices, such as bars welded to the piles, or with a
sufficient bond length to develop the applied uplift force.
11.3.5.7 Construction considerations—Piles should be
spaced close enough together to minimize the cost of the cap
or footing, and yet far enough apart, if friction piles, to mini-
mize the reduction in pile capacity caused by group action.
Other effects of closely spaced piles can be damage caused
by the driving of adjacent piles. This damage may be struc-
tural in nature, such as collapse of casings, splitting of
casings, permitting inflow of ground water, and damage to
green concrete. Such damage is usually remedied by
neglecting the supporting effect of the damaged pile and
replacing it with another pile. Close spacing may also cause
lateral displacement of adjacent piles and heave of
surrounding soil that lifts the already driven piles. In this case,
the affected piles are redriven to establish their capacity. Spec-
ifications should be written to cover these anticipated prob-
lems. Consideration should also be given to various methods
of installation, and whether it is practical to obtain required
penetrations with the type of pile being evaluated.
11.3.6 Special types
11.3.6.1 Caissons—When the subsoil at the pier loca-
tion consists of a deep strata of sand that has a possibility of
scour to a significant depth, or when deep penetration
through overlying soils is required, the use of piles or drilled
piers becomes expensive or impractical, and caissons may be
used. By definition, caissons are shells within which the
excavation is made.
Current practice is to construct caissons by the open-
dredge method. The portion of the caisson below the bed of
the stream is a thick-walled concrete box with multiple cells
designed to function as a permanent part of the foundation.
The portion of the caisson above the bed consists of steel
sheet piling which is removed after completion of the pier.
The concrete walls of the caisson are constructed thick
enough so that the caisson sinks under its own weight when
the soil within the caisson is excavated. All excavation is
done under water while maintaining a positive head of water
within the caisson. Control of the sinking is done by selective
excavation within the caisson cells. To insure control and
minimize drifting, the caisson should have at least four cells.
Upon reaching the proper founding depth, the bottom is
cleaned and sealed with a thick layer of concrete.
The water in the caisson is then pumped to an elevation
below the top of the concrete walls of the caisson, a footing
is constructed upon the walls, and the pier is completed.
 BRIDGE ANALYSIS AND DESIGN 343R-129

Fig. 11.3.6.1 illustrates a caisson constructed using a sand

island. Other methods are also used. A further description of
caissons can be found in References 11-5 and 11-31.
11.3.6.2 Other types—When special subsoil conditions
exist, foundations should be adapted to suit the conditions. In
most cases, these adaptations are merely variations of the
three usual types, i.e. spread, piles or drilled piers. When
subsoils change drastically from one pier location to the
next, consideration should be given to using different types
of foundations on adjacent piers.
Loose granular soil can be compacted or stabilized by
grouting or chemical injections to form a natural raft. Rock
anchors can be used to anchor footings resting on rock with
a sloping surface or bedding.
11.3.7 Special considerations
11.3.7.1 Cofferdams—When the pier site is dry during
most of the construction season, the excavation to the bottom
of the footing can usually be made by open excavation.
When adjacent construction precludes this, or when the
water level is or is expected to be above the ground surface,
the excavation is generally made within a cofferdam.
Cofferdams are dams with a brief service life, built around
the pier foundation so that construction operations can be
performed in the dry area below the adjacent water level. They
can be single wall, double wall, or cellular, braced or unbraced.
They are usually constructed of steel sheet piling, but have
been made of precast concrete, overlapping cast-in-place
concrete piles, or slurry walls. The portion of the cofferdam Fig. 11.3.6.1—Open-dredged caisson using sand island.
above the footing level is always removed, but the part below
may be left in place and, if anchored to the footing, may
provide additional protection against scour. References 11-32 and it is highly recommended as an excellent reference on
and 11-33 are recommended for guidance on this subject. the subject of bridge hydraulics.
When single wall steel sheet piling cofferdams are used in 11.4.2 Bridge location—The location and orientation of a
river piers, unless the pier is founded on rock, a concrete seal bridge relative to a stream channel is extremely important.
coat is required. The determination of that seal coat thickness The alignment of piers and abutments, as well as the
is generally the responsibility of the design engineer. The waterway provided, either exacerbate or ameliorate prob-
design criteria for seal coats may include, in addition to the lems at a bridge crossing. The selection of the location is an
weight of the seal coat itself, the weight of the steel sheet important first step in a successful bridge project.
piling, an allowance for friction between the sheet piling and 11.4.3 Waterway opening—Before a determination is
the surrounding earth, and the uplift resistance of the perma- made for the length and height of a bridge, the stream channel
nent piles, if any. The design plans should clearly state what should be studied for different stages of flow. Sometimes the
high water elevation was used in designing the seal coat and major portion of flood flow follows a quite different path of
should require ready means of flooding the cofferdam if this flow orientation than at low flow and provisions should be
elevation is exceeded. made for such an eventuality. The designer should determine
11.3.7.2 Impact during construction—Consideration what controls the stream cross section and how the flow is
should be made in the design of bridge piers for possible distributed throughout the flood plain. Crossing near and
accidental impacts during the life of the structure. It is also above confluences is often affected by backwater from the
prudent to consider the possible effect of an accidental other stream. The key to a good solution is to place the bridge
impact during the construction period. Piers under construc- waterway opening in the best location to accommodate all
tion with no superstructure load on them are less stable than stages of flow, but providing a waterway area does not neces-
piers carrying the superstructure. Specifications should call sarily mean that water will flow through it. Oftentimes the
the contractor’s attention to the need to protect against U.S. Geological Survey or a state agency has records or data
impacts caused by factors beyond the contractor’s control, available for determining the flow characteristics of a stream.
such as adjacent traffic, barge, or boat traffic. 11.4.4 Scour—The introduction of a bridge in a river
channel can change the natural stream environment and
11.4—Hydraulic requirements often causes scour and bank erosion with severe conse-
11.4.1 General—AASHTO Bulletin HDG-7 has been quences. A number of bridges have failed because of bed and
prepared by a task force of experienced hydraulic engineers, bank scour at piers and abutments. Extra precautions in
343R-130 MANUAL OF CONCRETE PRACTICE
regard to this phenomenon at the time of design and
construction will pay great dividends. Some of these precau-
tions may involve founding the piers on piles at river cross-
ings, instead of spread footings, and extensive rip-rapping of
the river banks.
Estimating depth and extent of scour is not a simple
matter. Research and field studies have been made, but expe-
rienced judgment is also needed. Piles, if used, should be
specified for penetration or tip elevation as well as their load
capacity. In some cases it may be necessary to assume that
scour lowers the river bottom below the pile footing and the
piles designed to function as free standing columns with a
lateral force acting on them.
11.4.5 Spur dikes—The spur dike, as used in bridge
construction, is a projection of earth or rock built integral
with the approach embankment, extending upstream and
downstream, to “guide” the flood flow under the end spans
of a bridge. The spur dike usually forms a portion of an
ellipse, and is sloped and shaped to minimize turbulence and
scour, especially when large quantities of flow are diverted
from a flood plain to the bridge opening. These dikes are
used extensively in the southern and central sections of the
U.S., where flood plains are quite wide. Cost effectiveness
dictates that the bridge structure cannot span the entire flood
plain, and the approach roadway is placed on an embank-
ment that becomes a wing dam during floods.
11.4.6 Slope protection—Approach roadway embankments
and spill-through abutments are vulnerable to attack by flood
water at most crossings. Some kind of protection is desirable,
but a grassed slope is often a simple solution, above the flood
line and rip-rap below. Excessive clearing of trees and brush
upstream from a bridge can cause the flow to accelerate and
aggravate scour at the bridge. However, the tendency to
remove trees and other obstructions has greatly decreased in
recent years due mostly to environmental concerns.
A source of available material for slope protection is
always of primary concern. Dumped rip-rap, consisting of
graded stone of sizes appropriate for the flow velocities and
depth of flow, is the most common type of protection. A
filter blanket of graded gravel or filter cloth is usually needed
beneath the stone rip-rap to prevent leaching of the under-
lying embankment material. Although the appearance of
hand-placed rip-rap is preferred by some engineers, its extra
cost is difficult to justify.
Slope paving under bridge ends is commonly a concrete
slab, sometimes jointed and reinforced with a wire fabric or
reinforcing steel. If care is not taken in providing a drainage
system and anchorage at the toe, a slope failure is likely to
occur. Such failures are very difficult to repair, making the
self healing stone rip-rap a superior solution. Concrete slope
paving is more commonly found under bridge ends that are
adjacent to roads and freeways.
11.5—Abutments
11.5.1 General—Abutments are the substructure elements
used at the end of the bridge to support the superstructure and
to retain the embankment that supports the approach roadway.
Generally, several types of abutments would be satisfac-
tory for a particular bridge site; economics is usually the
prime factor in selecting the type to be used. For river or
stream crossings, the minimum required channel width and
waterway area generally set the abutment location and type.
For highway overpasses, minimum horizontal clearances
and sight-distances should be maintained. An abutment
located at or near the top of a slope is not a collision obstacle,
but one located at the bottom may be. In this case, safety may
dictate. Esthetics is also a factor when selecting an abutment
type and location.
11.5.2 Loads and stability—Abutments should be
designed for horizontal loads due to earth pressure of the
approach roadway embankment, as well as the vertical loads
from the superstructure. They should be stable against over-
turning and sliding. The foundations of the abutment should
be designed to prevent excessive differential settlement and
lateral movement.
Vertical loads from the superstructure and dead load of the
abutment itself are generally well-defined, but consideration
should also be given to live and dead loads from the
approach roadway. These loads are carried by the abutment
footing as dead load plus live load surcharge, or by the abut-
ment backwall as the reaction of a reinforced concrete
approach slab. In all cases, the designer should consider not
only the effects of maximum loads, but also the effects of
minimum loads.
Horizontal design loads require a great deal of judgment
on the part of the engineer. Although the abutment functions
as a retaining wall, the horizontal movement of the top is
much more critical, and design earth pressures should reflect
this. The abutment is often restrained by the bridge super-
structure or by the wing walls from deflecting in the classical
manner that reduces the “at rest” pressure to active pressure.
Some designers increase the active pressure normally used
for retaining walls by 50 percent when designing abutments;
earth pressures also depend on the type of material used in
constructing the roadway embankment.
Piles installed in abutments that are set on fill are often
subject to excessive downdrag and sometimes lateral defor-
mation due to settlement within and under the fill. Therefore,
they should be installed through casings or sleeves with the
annular space filled with pea gravel.
Certain types of structures, such as rigid frame bridges or
short, single span bridges with fixed bearings at each abut-
ment, cause much greater soil pressures due to live load
action or temperature. Other types, such as “spill through”
abutments, require careful evaluation of earth pressures
acting on the piles, caissons, or columns supporting the abut-
ment cap. It is often assumed that the earth pressure acting
on these elements acts on a width twice as great as the actual
width of the element.
Abutments of bridges located in seismic zones are subject
to much greater forces; the AASHTO “Guide Specifications
for Seismic Design of Highway Structures” should be used
as a reference. To minimize potential loss of bridge access
arising from abutment damage, monolithic or end diaphragm
abutments are strongly recommended. Settlement or
 BRIDGE ANALYSIS AND DESIGN 343R-131

approach slabs should be provided and adequately linked to

abutments using flexible ties.
11.5.3 Types of abutments
11.5.3.1 Sill abutments [Fig. 11.5.3 (a) (b) and (c)]—
The sill abutment is constructed at the top of the slope after
the roadway embankment is close to final grade. Many
consider this abutment the best means of avoiding most of
the problems that cause rough riding approach pavements. It
eliminates the difficulties of obtaining adequate compaction
adjacent to the relatively high walls of closed abutments. In
addition, depending on the type of support used, the differ-
ential settlements of abutment and embankment under the
approach pavement can be minimized.
A berm is usually constructed at the front of the abutment.
The approach embankment may settle by forcing up or
bulging the slope in front of the abutment; the weight of the
berm helps prevent this.
11.5.3.2 Spill through abutments [Fig. 11.5.3 (d)]—
This type of abutment is frequently used where an additional
span is to be added to the bridge at a later date. It may also
be used to satisfy some unique construction problem. It is
supported on columns or stems that extend upward from the
natural ground, or it may be supported on drilled caissons or
piles that are constructed prior to placing the embankment.
It is very difficult to properly compact the embankment
materials that are to be placed around the columns and under
the abutment cap. Early settlement and erosion are problems
frequently encountered with this type of abutment.
If the abutment is to be used as a future pier, it is important Fig. 11.5.3—Abutment types.
that the wings and back wall be designed for each removal.
Construction joints are generally formed with preformed
joint material. Reinforcement bars are not extended through work near the walls, or if one end of the bridge is backfilled
the joints. Threaded inserts with bolts can be used to carry prior to the other; severe cracking may result. The placement
tension stresses across joints, depending upon the intent of of the embankment after abutment construction may cause
the design. foundation settlement. For these reasons, as much of the
11.5.3.3 Closed abutments [Fig. 11.5.3(e) (f) and (g)] roadway embankment as possible should be in place before
Full-closed abutments—A full-closed abutment is built starting abutment construction. Backfilling should be prohib-
close to the roadway or stream being crossed, and it retains ited until the superstructure is in place. Other disadvantages
the full height of roadway embankment. This type of abut- of full-closed abutments are minimum horizontal clearance,
ment is the most costly; however, it may be desirable where minimum sight distance, collision hazard, and settlement.
right of way is critical. By reducing the span length and Semi-closed abutments—The semi-closed abutment is built
superstructure cost, the total structure cost may be reduced. part way up the end slope of the roadway embankment, and it
Rigid frame structures use a full retaining abutment poured provides more horizontal clearance and sight distance than a
monolithically with the superstructure. If both abutments are full-closed abutment. Located on the embankment slope, it
connected by fixed bearings to the superstructure (as in rigid becomes less of a collision hazard for a vehicle out of control.
frames), the abutment wings are joined to the abutment walls The discussion about full-closed abutments generally
by a mortised expansion joint. For a nonskewed abutment, this applies to semi-closed types. They are used primarily in two-
enables the abutment wall to rotate about its base and allow for span highways over highway crossings as a substitute for a
superstructure contraction and expansion due to temperature shoulder pier and sill abutment. These abutments generally
and shrinkage, assuming the rotation is possible. It also allows are designed with a fixed base, allowing wing walls to be
differential settlement to occur between the wing and abutment rigidly attached to the abutment body. Wings and body are
footings. This differential settlement is not uncommon, due to usually poured monolithically.
the different loads on the two foundations. 11.5.3.4 Closed cellular abutments [Fig. 11.5.3(h)]—
An objectionable feature of full-closed abutments is the These abutments are also called vaulted abutments. Gener-
difficulty associated with placing and compacting material ally, these abutments are a combination of a solid pier (front
against the back of the abutment wall and between the wing wall) and a sill-type abutment, with precast concrete beams
walls. It is possible that this type of abutment may be shoved spanning between the front wall and the sill abutment. The
out of vertical alignment if heavy equipment is permitted to side walls act as curtains to hide what is in effect a short end
343R-132 MANUAL OF CONCRETE PRACTICE
span. Provisions should be made for access to the vault and
for ventilation.
Cellular abutments are often used where the main span
superstructure would require too tall a sill-type abutment.
They are also used to shorten the main spans, particularly on
bad skews, and the side walls provide an attractive esthetic
treatment for an otherwise awkward situation.
11.5.4 Retaining walls—Retaining walls, except for the
retaining function of abutments, are not considered a struc-
tural element of a bridge. Retaining walls are used to confine
soil so that roadways can be properly supported or protected.
Excellent aids for the design of retaining walls are the AREA
Manual, Section 5, Chapter 8.
Because retaining walls confining soils that support high-
ways or railroads have a much more critical role than typical
building retaining walls, greater care should be taken in their
design and detailing. Proper drainage of the soil being
retained is necessary.
11.5.5 Wing walls—Wing walls are retaining walls that
adjoin abutments. They may be separated from the abut-
ments, or they may be monolithic with the abutments. They
may have separate foundations, or they may be cantilevered
horizontally and vertically from the abutment, and they
generally support parapets that are designed for impact
loads. The tops of the parapets are usually sloped to prevent
vehicles from colliding with the end of the wing wall.
The junction of the wing wall with the abutment proper
requires special attention. The wing wall will tend to deflect
differently than the abutment, both in magnitude and direc-
tion. If the wing wall is monolithic with the abutment,
stresses at the junction will require special reinforcement.
These stresses are a combination of vertical and horizontal
bending and shear, plus torsion. If the wing wall is separate,
special joints and details are needed to prevent unsightly
gaps or offsets.
It is generally recommended that wing walls of closed
abutments be separated from the abutment proper, but wing
walls of sill-type abutments be monolithic. Wing walls may
be parallel to the road or stream being crossed, they may be
parallel to the roadway crossing the bridge, or they may be
flared. The choice is generally made on the basis of esthetics
and safety.
11.5.6 Joints at abutments—The ideal solution to the
roadway joint at the abutment is to eliminate it. In short span
concrete bridges, this can be done by making the deck slab
monolithic with the back wall of the abutment [Fig. 11.5.3(a)].
For bridges in the overall length range up to 300 ft (91 m),
the deck slab and girders may be cantilevered beyond the
back wall of the abutment and a diaphragm back wall (inte-
gral with the girders) placed against the soil [Fig. 11.5.3(b)].
These details effectively eliminate the expansion joint at the
end of a bridge.
For all other cases, an expansion joint should be placed
between the end of the deck slab and the back wall of the
abutment to permit unrestrained movement of the super-
structure. These joints are of many types; for a discussion of
the various types, see ACI 504R.
Another phenomenon that should be considered when the
approach roadways are concrete is the tendency for the pave-
ments to grow or lengthen, due to minute cracks being filled
with dirt during cold periods. This lengthening, if not prop-
erly controlled by adequate joints and maintenance, can
damage the abutment back walls. If a reinforced concrete
approach slab is used, a joint can be placed between the
approach slab and the abutment back wall, or it may be
located at the end of the approach slab away from the bridge.
11.6—Piers
11.6.1 General description—Bridge piers serve the
general purpose of transmitting vertical and horizontal loads
to the foundation. In addition to the requirement to carry
loads to the foundation, bridge piers often are required to
resist superstructure rotations, which are the result of
moments induced by frame action when piers are cast mono-
lithically with the superstructure.
Reinforced concrete is the material usually used in bridge
pier construction. Another type of pier utilizing concrete is a
composite column that is a combination of structural steel
and concrete. This is sometimes appropriate where steel
piles are utilized in a pile bent that needs to be increased in
section strength or to present a smooth wall for esthetics and
for protection against drift hangup. Also, it is possible that
site conditions might dictate that bridge piers be constructed
using precast concrete. This is best accomplished using
precast concrete segments post-tensioned together verti-
cally, and is covered in Section 11.6.5.
Often, bridge piers are the most decisive factor in
obtaining proper esthetics for a bridge structure. Proper loca-
tion, selection of a suitable configuration that blends well
with both the superstructure and the terrain, and careful
proportioning of section dimensions have a significant influ-
ence on structure appearance.
11.6.2 Pier configurations11-34—It is essential that pier
configurations be compatible with the type of superstructure
that they support, and also with the type of connection made
to the superstructure. Piers may be constructed to be integral
with the superstructure or they might be connected by a
pinned or an expansion-type of bearing.
Pier configurations may be single shafts or multiple
column bents. Single shaft piers generally blend well with
required esthetic treatments and offer minimal restriction to
drift in a stream. Multiple column bents are usually very effi-
cient from a structural viewpoint and are advantageous for
use in very wide structures, since joints in the cap can mini-
mize transverse shrinkage problems.
Generally, section shapes are either rectangular or circular,
unless some special consideration is given to architectural
treatment. With today’s increased concern for visual impact,
more consideration should be given to special shapes. Often,
suitable architectural treatment can be achieved with circular
or rectangular sections by providing items such as recessed
areas with brick veneer or a form liner.
Circular shafts are advantageous, based on the following:
a. Ease of forming, since many contractors have ready
access to steel forms.
 BRIDGE ANALYSIS AND DESIGN
b. Better flow characteristics, particularly when the
stream flow direction changes between low and flood
stage and where the river changes its course from
season to season.
c. It is never at a skew in a framed structure and therefore
avoids the problem of increased stiffness due to a pier
axis alignment that has a stiffness component in the
longitudinal direction of the superstructure.
d. Spirals or round ties make the placement of reinforce-
ment relatively easy. Spiral reinforcing is also efficient
in providing lateral support to vertical reinforcing. This
becomes important in seismic areas where the outer
cover of concrete may spall off due to severe dynamic
loadings. Properly designed spiral reinforcing provides
excellent ductility under seismic loading. Similar
ductility can be provided as needed in tied circular
columns by the use of closely spaced ties welded to
form a continuous loop.
Advantages of the rectangular shape are as follows:
a. It can be proportioned to place section strength in the
direction needed.
b. More reinforcement can be placed in areas of higher
strain to produce greater moment capacity.
c. For moments produced by change in length of the
superstructure between two fixed piers, rectangular
shafts have a greater moment capacity versus moment-
generated ratio than do circular shafts.
Fig. 11.6.2(a) and (b)—Hammerhead and solid wall piers.
343R-133
d. Rectangular wall-type piers offer little resistance to
stream flow, but only if they are parallel to the stream
flow.
Fig. 11.6.2(a) to Fig. 11.6.2(h) from Reference 11-34
show typical pier configurations and approximate dimen-
sions for use in initial design. Such dimensions should be
refined in final design. These figures also provide some
direction as to what configurations are suitable for various
heights of substructure.
For river piers, configurations should be based on consid-
eration of flow characteristics. The alignment of a shaft
should be made, as close as possible, to the direction of
stream flow at higher stages. Circular ends of rectangular
shafts that are closely aligned with the stream flow will
provide flow characteristics similar to circular shafts. There-
fore, if the stream flow pressure equation (Section 5.5) SF =
KV2 is used to calculate pressure, a value of K = 0.67 is suit-
able for use.11-13
In streams carrying large ice floes, a configuration should
be selected that will pass the floe without forming dangerous
ice jams. A thick sheet of moving ice striking a pier could do
several things; it could pivot about the pier, fail in one of
several possible mechanisms, or come to rest between two
piers, or a pier and adjacent bank creating a potentially
dangerous situation. Sloping nose piers shown in Fig.
11.6.2(i) serve to break up such ice sheets and reduce the
force. When a sheet of ice impinges on a sloping pier nose,
Fig. 11.6.2(c) and (d)—Circular shaft hammerhead and
multicolumn piers
343R-134 MANUAL OF CONCRETE PRACTICE

Fig. 11.6.2(h)—Multistory bent piers.

Fig. 11.6.2(e) and (f)—Multicolumn piers and tapered solid

wall piers.

Fig. 11.6.2(g)—Voided box section piers. Fig. 11.6.2(i)—Piers with ice breaker.
 BRIDGE ANALYSIS AND DESIGN 343R-135

the sheet tends to ride up on the nose, and when a sufficient d. Specially designed connections (Fig. 11.6.3.2) can
area is supported above the water, the sheet fails in shear and produce a frame moment restraint at the top and bottom
flexure. Calculations for design ice pressure are given in of a pier shaft along with a direct load transmitted
Section 5.5. through the bearing. This type of connection is
11.6.3 Connections to the superstructure—Types of connec- common with precast cantilever superstructure
tions of superstructure to piers should be consistent with the construction.
type of superstructure. In addition, when selecting a type of 11.6.3.3 Articulated hinges—Articulated hinges are an
connection, consideration should be given to the desirability of efficient way to reduce the effect of pier stiffness, when such
moment connections, potential uplift, seismic motion, magni- stiffness exceeds that which is structurally efficient.
tude of potential lateral forces, and other such items. The Details are shown in Figs. 11.6.3.3(a) to (f). Note that the
following are types of connections to be considered. stiffness is reduced with a hinge at the bottom of the column
11.6.3.1 Monolithic connections—By definition, this (b). With the articulated hinge at the top of the column (c),
type of connection is one where the concrete for the super- all moment stiffness is eliminated. If an articulated hinge is
structure and the substructure is cast together to act as a provided top and bottom of the column (d), then an expan-
framed unit, where moments are transferred through the sion condition is created where structural stability is
connection from the superstructure to the substructure and provided by other piers or abutments in the structure.
vice versa [Fig. 11.6.3.3(a) and (b)]. Construction joints are 11.6.4 Design considerations
normally provided between the pier and the superstructure in 11.6.4.1 General—All bridge piers and compression
order to minimize problems due to shrinkage of the fresh members should be designed by strength design procedures
concrete in the pier shaft form. Construction joints are also with appropriate consideration given to serviceability at a
sometimes provided in the superstructure to reduce working stress level. The primary serviceability criteria
shrinkage and creep effects acting on the piers as a result of should be a crack control evaluation. The recommendations
the superstructure shortening between two framed piers. given in Section 8.4 should be followed.
Vertical reinforcing in the piers should be extended a devel- The capacity analysis should be based on a concrete ulti-
opment length past the pier superstructure connection in order mate compressive strain of 0.003 and should consider stress
to provide sufficient moment strength in the pier shaft. and strain compatibility for both direct load and moment.
The strength for pure compression, balanced conditions, and
If the pier is equal in width to the out-to-out spacing of the
pier bending should be as given in Chapter 7.
main superstructure load carrying members, the entire shaft
11.6.4.2 Slenderness—Slenderness considerations
cross section may be extended into the superstructure. Where
involve control of the slenderness ratio kl/r and computation
the width of the superstructure extends beyond the plan
of secondary moments at the ultimate state.
dimension of the column, pier caps should be provided and
designed to transmit the longitudinal moments by torsion.
The desirability of using a monolithic pier superstructure
connection depends on relative pier stiffness. A framed pier
that is properly proportioned will increase support restraints
sufficiently to substantially reduce the overall moment
requirement in the superstructure. Piers that are short and
correspondingly stiff will cause very high moments due to
changes in length of the superstructure (elastic shortening,
temperature, shrinkage, creep effects) and therefore, be very
inefficient. In cases such as this, it is recommended to
provide sliding bearings at the top of all but one of the piers
to eliminate these undesirable moments.
11.6.3.2 Bearings—The subject of bearings and bearing
design is discussed in Section 11.2. In considering the effect
of bearings on pier design, the following can produce loads
and moments to be considered in pier design:
a. Transverse or longitudinal shear restraints in the bear-
ings can cause pier shaft bending.
b. Lateral loads, applied some distance above the pier top,
can introduce a moment at the top of the pier when such
loads are applied in the transverse direction against a
series of bearings (this moment is transmitted to the
pier by the variation of bearing loads).
c. Vertical loads on bearings not only produce direct load
in a pier shaft but also moments when bearings are Fig. 11.6.3.2—Special moment connections of superstructure
placed eccentrically. to pier.
343R-136 MANUAL OF CONCRETE PRACTICE

and the critical buckling load Pcr. The factor is given to

convert any column or pier into an equivalent pin-ended
column bent in single curvature. The variation in effective
length factor can be considerable, depending upon end
conditions and the braced or unbraced condition.
The effective length factor can be selected from the charts
shown in ACI 318R. The chart for unbraced frames should
be used. It may also be determined by the following equa-
tions for braced piers11-38

k = 0.7 + 0.05 ( G A + G B ) ≤ 1.0 (11-1)

k = 0.85 + 0.05 ( G min ) ≤ 1.0 (11-2)

Use the smaller of the two values.

Use the following equations for unbraced piers11-39

20 – G avg
For G avg ≤ 2k = ----------------------
- 1 + G avg (11-3)
20

For G avg ≥ 2k = 0.9 1 + G avg (11-4)

where

EI
------ Columns
L
G A = -------------------------------------------------------------------------
EI Members resisting column
------
L bending at A end of column
Fig. 11.6.3.3(a) to (f)—Monolithic and hinged connections.
In computing effective length factors for monolithic
In most cases, kl/r should be limited to a maximum of 100. connections, it is important to properly evaluate the degree
The effective length value k is computed as discussed in of fixity in the foundation. The following values can be used:
Section 11.6.4.3, and the radius of gyration r is computed as GA
0.25 times the diameter of a round column and 0.30 times the 1.5 Footing anchored on rock
depth of a rectangular column. If kl/4 exceeds 100, a precise 3.0 Footing not anchored on rock
secondary analysis should be made that includes the influ- 5.0 Footing on soil
ence of axial loads and variable moment of inertia on 1.0 Footing on multiple rows of piles
member stiffness and fixed end moments. This analysis In determining these effective length factors, the designer
should also include the effects of creep due to long term should use some judgment as to whether there is a braced or
loads, along with the effect of deflections on moments and an unbraced condition. It is reasonable to assume that most
forces. For columns with a kl/r value less than 22, secondary bridge analysis should use the unbraced condition.
effects might be ignored. 11.6.4.4 Biaxial bending—All pier section analysis
Secondary effects should be computed using a secondary should be done using strength design procedures. For biaxial
analysis as previously suggested, or by using an approximate bending, either an actual stress and strain compatibility anal-
method for developing secondary moment effects. Several ysis should be made or an approximate analysis that will
approximate methods are appropriate. The moment magni- simulate such.
fier given in ACI 318 and explained in ACI 318R is the most As an approximate analysis when Pu > 0.1fc′Ag, the recip-
straightforward approach and usually is the most practical rocal load equation
procedure for making this analysis. As an alternative, it
would be acceptable to use the P-Delta method, as described 1/Pnxy = 1/Pnx + 1/Pny + 1/Po (11-5)
in References 11-35, 11-36, and 11-37, which may result in
reduced amounts of reinforcement in the pier over the should be used. This equation generally gives conservative
previous simplified analysis. results and is practical to use.
11.6.4.3 Effective length factors—The effective length In reality, most bridge piers will be subjected to ultimate
factor k is used in computing both the slenderness ratio kl/r loads where Pu < 0.1fc′Ag. The elliptic equation
 BRIDGE ANALYSIS AND DESIGN
(Mux/φMnx)2 + (Muy/φ/Mny)2 ≤ 1.0 (11-6)
gives a reasonably good simulation of the stress and strain
compatibility analysis.
11.6.4.5 Irregular shapes—The capacity analysis for
irregular and unsymmetrical shapes is somewhat similar to
the capacity analysis for biaxial bending. However, simple
approximate solutions are not readily available. A stress and
strain compatibility analysis is suitable for determining the
capacity of these unusual shapes. This analysis is most suit-
ably carried out by using a computer.
In the analysis of many irregular shapes, consideration should
be given to using an increased concrete strain of 0.004 at the
ultimate load provided, however, the strength of the section
where the strain is greater than 0.003 is not included in the
computation of the required strength of the member.
In shapes where thin walls exist, it is important that these
thin portions of the section be analyzed for shear capacity
and requirements. This is particularly important when
considering large dynamic loads due to seismic action.
In areas where a detailed seismic analysis is required, the
columns theoretically should be designed for the maximum
compressive load due to gravity and seismic loading acting
simultaneously on the column. However, there is little guid-
ance available for computing the vertical seismic load. In
general, it would not significantly change the results if the
vertical seismic load was ignored.
11.6.4.6 Tie requirements—It has been generally
accepted that up to a point near the crushing strain of
concrete, lateral reinforcement does little to enhance the
structural performance of a column. Beyond this point,
however, the tied column is liable to exhibit a brittle failure
due to the fact that ties are normally widely spaced, and after
any concrete deterioration, there is inadequate support for
the main vertical reinforcing against buckling and for the
concrete core against crushing. Spiral columns usually have
spirals at a much closer spacing, and these perform better
when the concrete starts to deteriorate. This normal close
spacing of spirals is adequate to restrain the vertical reinforcing
against buckling.
Spiral columns also have the added benefit that the spiral
reinforcing creates a hoop tension confining concrete, and
thereby increasing the strength and ductility of the concrete
core. Closely spaced ties, whose ends are adequately
anchored inside the concrete core, will increase the strength
and ductility of tied columns, but this increase will not be as
significant as the same volume of closely spaced spirals. It is
not feasible to design a pier to withstand all damage from
severe seismic action. Therefore, current design practice is to
provide for ductility at potential plastic hinge areas, such as
in the piers, through increased hoops and ties or spirals.
It is recommended that the normal requirements for hoops
and ties be #4 hoops at 1 ft (300 mm) vertical spacing with
ties, not to exceed 2 ft (600 mm) transversely. As previously
suggested, these spacings should be reduced in potential
plastic hinge regions. Potential plastic hinge zones occur at
the bottom of piers for simply supported girder-type structures
343R-137
and at the top and the bottom of piers for framed-type struc-
tures. Ties should have hooks anchored in the compressive
zone (confined core).
The New Zealand Concrete Design Code gives guidance
for determining lateral reinforcing in the potential plastic
hinge regions. However, the recommendations of ATC-611-1
are rapidly gaining acceptance in the U.S. and should be used
where feasible. This code indicates that the length of the
potential plastic hinge region should not be less than a
column diameter for a circular column, or the longer column
cross section dimension for rectangular columns, or one-
sixth of the clear height of the column but not less than 18 in.
(450 mm) and, if applicable, this length should be increased to
cover the entire distance where the moment exceeds 0.8 times
the end moment. At no time should it be necessary for the
potential hinge length to extend over more than one-half of the
pier height. In situations where the maximum design load on
the pier exceeds φ(0.3)fc′Ag, the anticipated length taken from
the previous criteria should be increased by 50 percent.
In potential plastic hinge regions, it is recommended that
the maximum center-to-center spacing of transverse rein-
forcing not be less than the larger of one-fourth of the smaller
column dimension (column diameter for circular columns),
or 8 in. (200 mm).
In circular columns, ρs should not be less than the greater of
ρ s = 0.45 [ ( A g ⁄ A c ) – 1 ] ( f c ′ ) ⁄ f yh (11-7)
or
ρs = 0.12fc′ /fyh (11-8)
In rectangular columns, the total area of hoops and supple-
mentary cross ties in each of the principal directions of the
cross section within the spacing sh should not be less than the
greater of
Ash = 0.3shhc fc′/fyh[(Ag/Ac) – 1] (11-9)
or
Ash = 0.12shhc fc′/fyh (11-10)
In rectangular shafts, it is recommended that the center-to-
center spacing of tied bars not exceed the larger of one-third of
the column cross section dimension in that direction or 8 in.
(200 mm).
It is apparent that many piers that have wide cross sections
may have sufficient strength in the transverse direction of the
bridge to sustain seismic loads in the same direction without
yielding. This elastic response to a severe earthquake would
require a capacity to resist a lateral seismic load which is 4
to 6 times the usual code lateral seismic load. If such large
strength is available (both in bending and in shear), then it
would be only necessary to provide special detailing of rein-
forcement for horizontal loading in the longitudinal direction
of the bridge.
343R-138 MANUAL OF CONCRETE PRACTICE
Fig. 11.6.5(a) to (e)—Post-tensioned pier construction.
11.6.5 Post-tensioned piers—For very tall or highly
loaded piers, a nonprestressed reinforced pier may not be
practical. In such cases, post-tensioned piers may be used,
and these are usually segmental construction.
Precast segmental piers are also used in special cases
where particular construction requirements are specified.
Either the environment is fragile enough that the contractor
will have minimum access dictated by the locale or the
construction schedule, or the construction season is so short
that precasting of the piers is necessary to obtain a strict
construction schedule. A prime example of this is moun-
tainous regions, where the environmental constraints are
severe and the summer season is short. In such a case, the
pier segments can be precast during the winter and erected
the following spring. By precasting, intricate and interesting
shapes can be economical since the forms are reused. Each
segment is match cast to the previous segment to insure exact
seating when erected. Erection of a segmental pier, once the
foundation is cast, is simple. Using strands, both the founda-
tion and the shaft will require post-tensioning ducts placed
prior to pouring concrete [Fig. 11.6.5(a) and (b)]. Erection
continues by placing segment on segment [Fig. 11.6.5(c) and
(d)] until the desired height is obtained. Keys should be
placed in each section to aid in alignment. With all of the
segments in place, the pier is ready for stressing and then
grouting, followed by casting the pier cap [Fig. 11.6.5(c)
and (d)]. Joints between precast segments may be grouted
with cement or epoxy-injected. Joint edges should be sealed
against water penetration.
The weight of each individual segment should be kept at
approximately 40 tons (350 kN) to obviate the need for
special transport trucks or special overload permits. This
weight restriction will also help in erection, since most
bridge cranes can handle up to 40 tons (350 kN) for piers less
than 100-ft (30-m) high.
Generally, tensile stresses are not allowed under
AASHTO group loading conditions for the post-tensioning
design, tendons or bars may be used for the post-tensioning
system. If bars are used, an anchorage system in the footing
is required.
11.6.6 Detailing
11.6.6.1 Splices—Construction joints in piers, doweled
connections to footings, and reinforcing length limitations
often require splicing of the main vertical reinforcing within
the pier height. Reinforcing splices can be made by lap
splicing bars #11 and smaller, using mechanical splices, or
by welding. It is recommended that splices of adjacent bars
at the same vertical location be avoided, and vertical loca-
tions for splices be at least 3 ft (1 m) apart.
For lap splices where only alternate bars are spliced at the
same location, a Class “B” splice length, as defined in
Section 13.2, should be considered adequate. Proper clear
spacing between bars should be maintained in lap splice
areas, and mechanical splices should be detailed according
to the manufacturer's recommendations.
When welded splices are unavoidable, it is recommended
the splices be prepared in accordance with AWS D1.4. Care
should be taken during welding operations to protect adja-
cent bars from damage. Shielding devices are normally
provided for protection from splatter and errant contact with
a welding rod. It is also recommended that when Grade 60
bars are specified as welded, the reinforcement should be in
accordance with ASTM A 706.
11.6.6.2 Development requirements—When a pier shaft
has a moment resisting connection with a footing or with a
monolithic cap, full development of the vertical reinforcing
beyond the interface should be provided. This development
length may consist of a combination of the equivalent
embedment length of a hook or a mechanical anchorage, plus
additional embedment length of the vertical reinforcing. This
required additional embedment length will often dictate the
required size of footing. Pedestals may also be used in lieu of
increasing the footing depth.
If more vertical reinforcing is provided than actually
required by a capacity analysis, then the required develop-
ment length and the resultant required footing depth may be
reduced to provide only the capacity required.
11.6.6.3 Dynamic earthquake requirements—For
routine design, it is generally not cost-effective to design a
bridge structure to resist very large inertia loads resulting
from elastic response to severe earthquake action. Instead,
the design should be for smaller earthquake loads with the
structure being detailed to provide ductility in the piers.
Ductility becomes important when plastic hinge zones form,
since it gives significant energy dissipation. The value μ,
known as the ductility factor, is used to measure ductile action
and is defined as the ratio of the maximum displacement under
the design earthquake to the theoretical yield displacement.
An acceptable maximum design value of μ is 6.
As indicated in Section 11.6.4.6, adequate lateral pier rein-
forcement is most helpful in producing ductility. A general
 BRIDGE ANALYSIS AND DESIGN 343R-139

design procedure suggested for ductile structures that collapse of the bridge. Debris walls should provide smooth,
includes plastic hinge areas in piers should be: exposed faces capable of deflecting debris through the struc-
a. Design plastic hinge sections to have a dependable ture. Normally, debris walls should extend 3 to 5 ft (1 to 2 m)
flexural strength which is at least equal to the required above the design high water elevation for the bridge and at
flexural strength. The dependable strength should be φ least 2 ft (0.6 m) below the probable scour elevation.
times the ideal strength, based on the specified material 11.7.3 Crash walls—Railroads passing under bridge struc-
strengths of the steel and concrete. tures that use multiple column piers will often require crash
b. Design all sections, other than plastic hinge areas, for walls either in front of or integral with the piers that are adja-
flexure and shear. This should be based on an analysis cent to the tracks. These crash walls should provide a smooth
using plastic hinge flexural capacities, based on over- surface to a passing train. On occasion, similar crash walls
strength materials and including an allowance for might be provided along with guardrail for multiple column
strength increase due to steel strain hardening. Design piers that are adjacent to highways with heavy truck traffic.
the plastic hinge areas for shear.
RECOMMENDED REFERENCES
11.7—Pier protection11-40,11-41 The documents of the various standards-producing organi-
11.7.1 Fender systems—In the case of bridge piers located zations referred to in this report are listed below with their
in the vicinity of navigation channels and subject to the serial designation, including year of adoption or revision.
danger of damage from ship collision, provision should be The documents listed were the latest effort at the time this
made to protect the piers from collisions. The need and cost report was written. Since some of these documents are
of fender systems can be avoided by using longer spans that revised frequently, generally in minor detail only, the user of
extend beyond physical channel limits. this report should check directly with the sponsoring group
In addition to such measures as adequate channel delinea- if it is desired to refer to the latest revision.
tion, warning lights on the structure, and other applicable
aids to navigation, consideration should be given to the American Association of State Highway and Transportation
construction of fenders at the piers. Officials
The amount of protection offered by the fenders should be FIM-2 Foundation Investigation Manual, 1978
based on a prediction of the probable magnitude of a colli- GSDB Guide Specifications for Seismic Design of
sion, taking into account the size and speed of vessels using Highway Bridges, 1983
the channel. HDG-7 Hydraulic Analyses for the Location and Design of
Fenders may take the form of actual physical barriers, Bridges
independent of the pier and isolating it completely from the
impact of a collision, or they may act as cushions, limiting American Concrete Institute
the magnitude of forces transmitted to the pier. In either case, 336.3R-72 Suggested Design and Construc-
they should be designed as energy absorbing devices. (Revised 1985) tion Procedures for Pier Founda-
Timber-rubbing planks may be placed vertically around tions
the fender system to deflect small boats; all timber used in 543.R-74 Recommendations for Design,
fender systems should be pressure treated. Manufacture and Installation of
Essential requirements for fender piling are as follows: Concrete Piles
a. High energy absorption capacity before rupture to 318-83 Building Code Requirements for
eliminate damages to the main structure. (Reapproved 1980) Reinforced Concrete
b. High internal damping. 318R-83 Commentary on Building Code
c. High fatigue strength. Requirements for Reinforced
d. Appreciable elastic movement. Concrete
e. Capability of absorbing inclined impacts and rubbing SP-53 Reinforced Concrete Structures in a
forces to eliminate damage to fendering. Seismic Zone, 1977
f. Combined with the main structure, sufficient static
resistance and mass to cause plastic deformation of the American Railway Engineering Association Manual for
ship's hull. Railway Engineering
Consideration might be given to prestressed concrete in Chap 8 “Concrete Structures,” Manual of Railway Engi-
the design of fender systems because of its excellent energy neering
absorption. Prestressed sections should be properly confined
with spirals or ties. ASTM International
11.7.2 Debris walls—Multiple column piers located in STP 444 Performance of Deep Foundations, 1969
areas of streams where drift is anticipated, should have a STP 450 Vibration Effects of Earthquakes on Soils and
debris wall between the columns to minimize the possibility Foundations, 1969
of debris hangup. Debris hangup on piers can, in some STP 670 Behavior of Deep Foundations, 1979
instances, restrict the waterway opening, cause extreme STP 399 V-Average Shear and Cone Penetration Resis-
flooding upstream of the bridge and, in extreme cases, cause tance Testing of In-Situ Soil, 1966
343R-140 MANUAL OF CONCRETE PRACTICE
STP 412 Use of Nuclear Meters in Soil Investigations,
1968
STP 479 Special Procedures for Testing Soil and Rock
for Engineering Purposes, 1970
A 706-84a Low-Alloy Steel Deformed Bars for Concrete
Reinforcement
D 1143-81 Method of Testing Piles under Static Axial
Compressive Load
American Welding Society
D1.4-79 Structural Welding Code—Reinforcing Steel
Concrete Reinforcing Steel Institute
CRSI Handbook, 1984 (6th Edition)
Transportation Research Board
NCHRPR 248 Elastomeric Bearings Design, Construc-
tion and Material, 1982
REC 982 Bridges and Foundations, 1984
NCHRP SYN 107 Shallow Foundations for Highway
Structures, 1983
CITED REFERENCES
11-1. “Seismic Design Guidelines for Highway Bridges,”
ATC-6; Applied Technology Council, 1983.
11-2. Lambe, T. W., and Whitman, R. V., Soil Mechanics,
John Wiley & Sons, Inc., New York, 1979.
11-3. “Design Manual—Soil Mechanics, Foundations,
and Earth Structures,” U.S. Department of the Navy, Naval
Facilities Engineering Command, NAVFAC DM-7.2,
Washington, D.C., May 1982.
11-4. Zeevaert, L., Foundation Engineering, Van
Nostrand Reinhold Co., New York, 1985.
11-5. Tschebotarioff, G. P., “Foundations, Retaining and
Earth Structures,” McGraw-Hill Book Co., New York, 1973.
11-6. Terzaghi, K., and Peck, R. B., “Soil Mechanics in
Engineering Practice,” Second Edition, John Wiley and
Sons, New York, 1967.
11-7. Teng, W. C., Foundation Design, Prentice-Hall,
1962.
11-8. Jumikis, A. R., Foundation Engineering, Intext
Educational Publishers, 1971.
11-9. Dunham, C. W., Foundations of Structures,
McGraw-Hill Book Co., Second Edition, New York, 1962.
11-10. Leonards, G. A., et. al., Foundation Engineering,
McGraw-Hill Book Co., New York, 1962.
11-11. Parcher, J. V., and Means, R. E., Soil Mechanics
and Foundations, C. E. Merrill Publishing Co., 1968.
11-12. Tomlinson, M. J., “Foundation Design and
Construction,” Second Edition, Wiley Interscience, A Divi-
sion of John Wiley and Sons, Inc., New York, 1969.
11-13. Bowles, J. E., Foundation Analysis and Design,
McGraw-Hill Book Co., Third Edition, New York, 1982.
11-14. Hammond, R., Modern Foundation Methods,
Mclaren, London, 1967.
11-15. Carson, B. A., “Foundation Construction,”
McGraw-Hill Book Co., New York, 1965.
11-16. McClure, G. S., and Herman, G. J., “A Study of Ice
Formations and Their Effects on Bridges,” Highway
Research Report, Montana State Highway Commission,
Apr. 1965.
11-17. “Subsurface Investigations for Design and
Construction of Foundations of Buildings: Part 1,” American
Society of Civil Engineers, Journal of the Soil Mechanics
and Foundations Division, V. 98, No. SM5, May 1972, pp.
481-490.
11-18. “Subsurface Investigations for Design and
Construction of Foundations of Buildings: Part II,” Amer-
ican Society of Civil Engineers, Journal of the Soil
Mechanics and Foundations Division, V. 98, No. SM6, June
1972, pp. 557-575.
11-19. Chellis, R. D., Pile Foundations, McGraw-Hill
Book Co., New York, 1961.
11-20. “Behavior of Piles,” Proceedings of the Conference
Organized by the Institution of Civil Engineers in London,
Sept. 15-17, 1970, The Institution of Civil Engineers,
London, 1971.
11-21. Li. St. T., and Liu, T. C., “Prestressed Concrete
Piling—Contemporary Design Practice and Recommenda-
tions,” Journal of the American Concrete Institute, V. 67,
No. 3, Mar. 1970, pp. 201-220.
11-22. Gerwick, Jr., B. C., “High Capacity Prestressed
Concrete Piling,” Proceedings of the International Conference
on Planning and Design of Tall Buildings, Technical
Committee No. 11, Discussion No. 1, V. 1A, 1972, pp. 985-989.
11-23. Anderson, A. R., and Moustafa, S. E., “Dynamic
Driving Stresses in Prestressed Concrete Piles,” Civil Engi-
neering, ASCE, V. 40, No. 8, Aug. 1971, pp. 55-58.
11-24. Anderson, A. R., and Moustafa, S. E., “Ultimate
Strength of Prestressed Concrete Piles and Columns,”
Journal of the American Concrete Institute, V. 68, No. 8,
Aug. 1970, pp. 620-635.
11-25. Li, S. T., and Ramakrishnan, V., “Optimum Design
of Prestressed Concrete Combined Bearing and Sheet Piles,”
Journal of the Prestressed Concrete Institute, V. 17, No. 5,
Sept.-Oct. 1972, pp. 39-48.
11-26. Li, S. T., and Ramakrishnan, V., “Optimum
Prestress, Analysis, and Ultimate Strength Design of
Prestressed Concrete Bearing Piles,” Technical Report
prepared for the Engineering Division, Civil Works, Office
of the Chief of Engineers, Dept. of the Army, Washington,
D.C., 1969.
11-27. Li, S. T., and Ramakrishnan, V., “Optimum
Prestress, Analysis, and Ultimate-Strength Design of
Prestressed Concrete Combined Bearing and Sheet Piles,”
Technical Report prepared for the Engineering Division,
Civil Works, Office of the Chief of Engineers, Dept. of the
Army, Washington, D.C., Apr., 1970.
11-28. Bruce, R. N., and Hebert, D.C., “Splicing of Precast
Prestressed Concrete Pile: Part I—Review and Performance
of Splices,” Journal of the Prestressed Concrete Institute, V.
19, No. 5, Sept.-Oct. 1974, pp. 70-97.
11-30. Li, S. T., and Ramakrishnan, V., “Ultimate Energy
Design of Prestressed Concrete Fender Piling,” Journal of
the Waterways, Harbors and Coastal Engineering Division,
ASCE, V. 98, No. WW4, Nov. 1971, pp. 647-662.
 BRIDGE ANALYSIS AND DESIGN
11-31. Peck, R. B., Hanson, W. E., and Thornburn, T. H.,
Foundation Engineering, Second Edition, John Wiley and
Sons, Inc., New York, 1974.
11-32. Bradley, J. N., “Hydraulics of Bridge Waterways,”
U.S. Federal Highway Administration, Design Series No. 1,
1970.
11-33. “Guide to Bridge Hydraulics,” University of
Toronto Press, Canada, 1973.
11-34. Heins, C. P., and Lawrie, R. A., “Design of Modern
Concrete Highway Bridges,” John Wiley & Sons, Inc., New
York, 1984.
11-35. Wood, B. R.; Beaulieu, D.; and Adams, P. F.,
“Column Design by P-Delta Method,” ST 2, Journal of the
Structural Division, ASCE, Feb. 1976.
11-36. Wood, B. R.; Beaulieu, D.; and Adams, P. F.,
“Further Aspects of Design by P-Delta Method,” ST 3,
Journal of the Structural Division, ASCE, Mar. 1976.
343R-141
11-37. MacGregor, J. G., and Hage, S. E., “Stability Anal-
ysis and Design of Concrete Frames,” ST 10, Journal of the
Structural Division, ASCE, Oct. 1977.
11-38. Cranston, W. B., “Analysis and Design of Rein-
forced Concrete Columns,” Research Report 20, Paper
41.020, Cement and Concrete Association, London, 1972,
54 pp.
11-39. Furlong, R. W., “Column Slenderness and Charts for
Design,” ACI JOURNAL, Proceedings V. 68, No. 1, Jan. 1971,
pp. 9-17.
11-40. “Bridge Pier Ship Collision,” National Research
Council, Marine Board Report, Jan. 1984.
11-41. “Criteria for the Design of Bridge Piers with
Respect to Vessel Collision in Louisiana Waterways,”
prepared for the Louisiana Department of Transportation
and Development and the Federal Highway Administration
by Modjeski and Masters, Consulting Engineers, Nov. 1984.
343R-142 MANUAL OF CONCRETE PRACTICE
Construction of precast concrete box girder bridge (photo courtesy of Texas Highway Department).
CHAPTER 12—PRECAST CONCRETE
12.1—Introduction
12.1.1 General—Precast concrete, manufactured either at
a plant or at the bridge site, offers many potential advantages
in quality control, speed of construction, and frequently,
economy.12-1—12-5 Precast concrete bridges, both short and
long spans, have been built in many environments, from
highly urbanized to rural.
Precast concrete bridge elements are generally joined with
other precast or cast-in-place elements to perform as if
constructed monolithically.12-1 Precast concrete components
may be fully or partially prestressed, or they may be rein-
forced only with conventional reinforcement.
Standardized shapes for highway bridge girders and slabs,
developed by a joint AASHTO-PCI committee, have been
widely used for span lengths up to 160 ft (49 m) (PCI STD-
101, 107, 108, 114 and 115). Special shapes for short-span
railway bridges have also been standardized by a joint
AREA-PCI committee (AREA, Chapter 8). Precast
substructure elements such as piles have been used for many
years; AREA Committee 8 has standardized precast concrete
caps for precast concrete piles.
Precast concrete elements are usually produced in well
established precasting plants. Typically, those plants employ
reusable forms. Considerable regional differences exist in
the readily available shapes. In fact, from state to state, large
variations exist in the products and shapes customarily used.
Therefore, in the preliminary planning stage, depending on
the scope of the project, it may be advisable to contact local
precasters to determine which shapes are readily available.
12.1.2 Advantages and limitations—Precasting offers
favorable conditions for forming (ease of building and strip-
ping), placement of reinforcement, and placing and curing of
concrete because the casting is done in a controlled environ-
ment. Precasting can be particularly advantageous in adverse
environmental conditions, such as heavy precipitation,
freezing, or hot, dry weather. Precasting also facilitates
rejection of a member which does not meet specifications.
Precast concrete bridge construction methods frequently
offer substantial economical savings because on-site labor
can be minimized, formwork supports eliminated, erection
performed during more favorable seasons and conditions,
labor and equipment utilized with maximum efficiency, and
advantage taken of mechanization in manufacture. Higher-
strength materials are often used with precast concrete,
permitting the use of thinner sections, which results in a
savings in material.
Precast concrete members can readily be erected over
existing highways, railroads, waterways, deep canyons, and
at other locations where the use of falsework to support
forms for cast-in-place concrete is impractical or undesir-
able. Erected precast beams may support formwork for a
 BRIDGE ANALYSIS AND DESIGN
subsequent cast-in-place concrete deck. Special construction
methods such as balanced cantilever, spliced girders, span-
by-span progressive placing, incremental launching, canti-
lever and suspended span, or stayed girder may be applicable
under special circumstances.
Railway companies utilize precast concrete frequently for
short spans because it is economical and can shorten
construction time drastically. As a rule, railroads procure
standard precast concrete components in large quantities and
transport them by rail to their own storage yards. When a
bridge is to be built, a special construction train takes all
materials, bridge members, and construction equipment to
the bridge site.
Optimum use of precasting is achieved by selecting it in
the preliminary design phase. If precast concrete elements
are substituted into a structure, with details drawn as a cast-
in-place construction, full advantages of precasting the
elements may not be realized.
The size of precast segments may be limited by economic
and practical considerations. Designers should evaluate struc-
tural lightweight concrete and hollow and thin sections made
of high-strength concrete as a means of reducing the weight of
precast concrete segments. Weight reduction facilitates more
economical transportation and erection of the elements.
In selecting the size and shape of precast bridge members,
site conditions and the capabilities of local contractors
should be considered. When feasible, contractors should be
allowed to submit an equivalent precast alternate.
12.2—Precast concrete superstructure elements
A wide variety of precast elements are available for short
span bridges (see subsections 4.9.9, 4.10.1, 4.10.3, 4.10.5,
Fig. 6.4.1, including References 12-3 and 12-6). Most
precast elements are standard shapes made with reusable
forms, which are carried in stock by precasting plants. For
medium and long-span bridges, special elements may be
more suitable.
12.2.1 Standard pretensioned concrete I-beams—The
most common precast bridge members are the AASHTO-
PCI Standard Pretensioned I-beams.12-2,12-4 Many states
have their own unique shapes, and most of them contain less
concrete than comparable AASHTO beams. In 1988, the
new PCI standard bulb-tees (BT-54, BT-63 and BT-72)
contained in PCI STD 115-87 were accepted by AASHTO.
They are now preferred for spans of 70-150 ft (21-46 m) by
some states.
Generally, precast pretensioned I-beams are longitudinal
beams supporting a composite cast-in-place deck [see
subsection 4.10.3 and Figs. 2.5.3.2 and 6.4.1(c)]. This
construction is usually continuous for live load, as
mentioned in sections 9.14.3 and 10.2.4.3 and References
12-7 and 12-8. Spans for this type of construction range from
30 to 160 ft (9 to 49 m).
12.2.2 Precast pretensioned deck panels—Thin precast
deck panels, serving as stay-in-place forms for a cast-in-
place deck, are sometimes used in conjunction with the
precast pretensioned I-beams. Deck panels may be saw-cut
diagonally to fit the ends of skew bridges. Careful attention
343R-143
Fig. 12.2.3.1—Precast trapezoidal channel.
must be paid to the bearing of precast panels on the tops of
the supporting beams (Reference PCI JR-343).
12.2.3 Precast trapezoidal concrete beams—Precast trap-
ezoidal channel sections, developed in Canada, are shown in
Fig. 12.2.3.1. They also require a cast-in-place deck and are
suitable for spans of 100 to 150 ft (30 to 45 m).12-1
12.2.4 Complete precast superstructures—Complete
precast concrete superstructures, requiring no cast-in place
deck, are used frequently.12-6 The precast units serve as the
supporting members, as well as the deck. Single tees, double
tees, tri-tees, multi-stem sections (also known as rib deck),
voided slabs (or hollow core slabs) shown in Fig. 2.5.3.1,
bulb tees, butted box beams [Fig. 6.4.1(e)], and inverted
channels fall in this category. A serious drawback of the
smaller box beams is the fact that they can not be inspected
from the inside by ordinary means. When superstructure
units deteriorate, epoxy repairs may be contemplated, but
usually entire members have to be replaced.
Superstructure units are placed side by side with a longitu-
dinal keyed joint between units. Inverted channels, box
beams, and some of the other units may either be convention-
ally reinforced or pretensioned. Pretensioned units are ordi-
narily furnished with square ends, but the various tee-shapes
and box beams can easily be made with skew ends. Nonpre-
stressed units, mostly inverted channel units, are usually
made to fit the skew, as shown in Fig. 12.2.4.1. When units
having square ends (normal units) are used in a skew bridge,
they are arranged as in Fig. 12.2.4.2.
Due to variations in concrete and prestressing forces, the
camber of pretensioned concrete superstructure units may
not be uniform. Unexpected changes in camber may occur
when the units are hauled long distances and over rough
roads. This can be an important factor when units are deliv-
ered to sites in the mountains or in remote areas. Bowing and
warping of such pretensioned units may result from
improper storage, transportation, rough handling, etc.
Thus, the tops of pretensioned channels and boxes may not
line up well enough to provide a smooth riding surface. On
343R-144 MANUAL OF CONCRETE PRACTICE
minor unpaved or low-volume roads, a gravel or asphalt
topping, underlain by a waterproof membrane, can serve as
a riding surface. Likewise, an allowance must be made for
horizontal misalignment. Substructure dimensions should
allow for extra superstructure width; a common allowance is
1/2 in. (13 mm) for each longitudinal joint.
Special attention is required for fastening these units
together, otherwise maintenance problems will arise after only
a few years. Load transfer across longitudinal joints is usually
accomplished by continuous shear keys, filled with nonshrink
grout, combined with tie bolts and/or welded connections.12-
9,12-10 part 6
As an alternate, transverse post-tensioning
combined with longitudinal continuous grout keys between
units may be used.12-11 This later alternative often includes a
2-in.- (50-mm-) thick high-density concrete overlay.
To obtain a better, more durable structure, a composite
reinforced concrete slab, 4-in. (100-mm) or thicker, can be
cast on top of the precast units. Extra transverse reinforcing
steel should be provided in the cast-in-place slab across the
longitudinal joints between precast units. Alternatively,
transverse post-tensioned tendons can be installed just below
the top of the units, as previously mentioned.12-11
An example of the advantages of precast concrete super-
structure units is the 5900-ft- (1.8-km-) long bridge carrying
Highway US 95 across Lake Pend Oreille near Sandpoint,
Idaho (see Fig. 12.2.4.3). Except for one 83 ft (25 m) navi-
gation span, the trestle contains 154 spans of 35 ft (11 m),
interspersed with 25 braced spans of 17 ft (5 m). The super-
structure is made up of 6-ft- (2-m-) wide pretensioned
Fig. 12.2.4.1—Plan view of skew bridge with skew units.
Fig. 12.2.4.2—Plan view of skew bridge with normal units.
concrete rib-deck units. Pretensioned box girders in the navi-
gation span weigh 32 tons (29 metric tons) each.
The entire deck was overlaid with a 4 in. (100 mm) cast-
in-place concrete topping. All pile caps and superstructure
units were manufactured in Spokane, Washington, and
hauled 75 miles (120 km) to the bridge site. The structure
was completed in 1981.
12.2.5 Precast concrete slabs for redecking—Existing
bridges—Precast concrete deck slabs have been used for
replacing worn-out bridge decks, particularly when the
bridge cannot be closed to traffic during rehabilitation. Such
deck slabs are usually pretensioned.
12.3—Segmental construction
12.3.1 General—Main longitudinal elements, comprising
a partial or complete transverse cross section of a bridge,
may be precast in lengths shorter than the span. Such shorter
elements are erected and prestressed together longitudinally
to act as an integral unit. These shorter elements are an
example of transverse segments as referred to in section
12.1. This method is known as segmental construction.
Segmental construction is most commonly used in precast
box girder bridges (see “AASHTO Guide Specifications for
Design and Construction of Segmental Concrete Bridges”
Reference GSCB or subsection 4.10.6, 6.4.5 and Fig.
2.5.3.3). Segmental construction may also be used to extend
the length of precast pretensioned I-beam spans, as covered
in section 12.3.2. Transverse precast deck slab segments,
described below, are also segmental construction.
 BRIDGE ANALYSIS AND DESIGN
Fig. 12.2.4.3—Sandpoint Bridge over Lake Pend Oreille (photo: Will Hawkins).
As indicated previously, the manufacture of precast
concrete elements is particularly appropriate for trestle struc-
tures and viaducts, where many identical spans and pile
bents are needed. The bridge crossing Albemarle Sound near
Edenton, North Carolina, built in 1988, is an excellent
example of such construction. A major portion (78 percent)
of this 3.5-mile- (5.6-km-) long two-lane highway structure
is a low-level trestle, divided into 260-ft- (79-m-) long
continuous bridge units; the bridge units are segmental post-
tensioned slab bridges. The deck slabs are 34-ft-wide x 20-
ft- long (10.4 x 6.1 m), and they weigh 55 tons (50 metric
tons) each. Square precast prestressed concrete piling is
connected directly into the deck units. Two other superstruc-
ture alternatives and various substructure systems had been
included in the bid documents. As a result, the cost was cut
significantly.12-12
Another excellent example of post-tensioned precast
concrete construction is the Bahrain Causeway, completed
in 1986. This is the largest bridge project in the Middle East
(see Reference 12-13 for a detailed account of this impor-
tant project).
12.3.2 Spliced girder construction—In situations where a
full-span girder may be too long or too heavy to be shipped,
the precast girder may be fabricated in relatively short
segments. Those segments will then be transported to the
bridge site, where they are field-spliced to produce the full
girder length. Splicing may be done before or after the
girders are erected in their final position. Continuity at the
girder splices may be achieved by post-tensioning, conven-
tional reinforcement, or embedded structural steel shapes.
Bulb tees and AASHTO I-girders are the most common
sections in spliced girder applications.12-14
Segment size is selected to accommodate various
constraints: the span layout of a bridge is governed by site
conditions. The maximum practical girder length is
controlled by transportation limitations and facilities at the
precasting plant. Also, splices should be placed at locations
which are accessible at the construction site, or splice loca-
tions may be dictated by the flexural stresses in the girders.
343R-145
Splicing makes precast concrete a competitive construction
for span ranges of 150-250 ft (45-75m). Splicing can also be
economical for shorter spans of 100-150 ft (30-45 m).
12.4—Precast concrete substructures
Precast pretensioned concrete piling has been standard-
ized since 1963.12-2 A small pipe may be embedded along the
axis of the pile to facilitate jetting.
Precast concrete pier caps have been used in railroad and
highway trestle structures. Precast concrete pile caps usually
have large openings to receive the piling. A cast-in-place
concrete plug or grout is needed to connect the piling to the
cap. As an alternative, steel plates with stud anchors can be
embedded in the precast concrete cap beams. Pipe piles or H
piles can then be welded to the cap in the field. This elimi-
nates waiting for the grout to harden.
Precast columns and pier shafts may be constructed by
erecting precast segments vertically with prestressing forces
applied during and after completion of construction. Other
large substructure members, such as pier caps, which may be
too large to precast as a single unit, may be precast in
segments and post-tensioned together after erection. Abut-
ments made up entirely of precast concrete components are
being built by many railroad companies.
“Mechanically stabilized earth” is a generic term
describing reinforced earth structures. These are specialized
proprietary earth-retaining systems using precast reinforced
con-crete or metal-facing panels. Mechanically stabilized
earth systems have been used widely in the U.S. and Europe
to build highway bridge abutments. They are especially suit-
able in situations where large settlements are anticipated.
For many years, precast concrete caissons and floating-
box piers have been economical for large bridges spanning
over waterways.12-1
12.5—Design
12.5.1 General—All loading and restraint conditions from
manufacture to completion of the structure should be consid-
ered in the design of precast concrete systems. This includes
343R-146 MANUAL OF CONCRETE PRACTICE
Fig. 12.5.6.1—Wide beam with superelevated CIP deck.
form removal, yard storage, transportation, storage at the
site, final erection, and joining of the precast segments. If the
structure is to behave as an integral unit, the effects at all
interconnected and adjoining elements should be properly
evaluated.
Design of joints and connections should include the effects
of all forces to be transmitted, including those caused by
shrinkage, creep, temperature gradients, and variations in
ambient temperature, settlement, elastic deformation, wind,
earthquake, and erection loads, as well as dead and live
loads. Details should also be designed to provide for
adequate manufacturing and erection tolerances.
12.5.2 Erection requirements—Erection forces should be
treated as dynamic loads. Impact and unforeseen shifts in
load distribution should be taken into account. A liberal
factor of safety (5.0 is suggested) is appropriate when deter-
mining the load capacity of lifting devices.12-6,12-10
Anchorage of the lifting devices must be adequate to prevent
pull-out failure. The lifting forces may be applied either to
the member at a specific angle, or they may be effective over
a range of angles. Consideration should also be given to
sway or swing of the component, which can put additional
strain on the lifting devices and may cause local concrete
crushing. Stability problems caused by the imposed lifting
forces and bending moments should be considered.
12.5.3 Handling precast units—Prestressed concrete units
are generally sensitive to positions other than the final
erected position, and temporary tensile stresses should be
evaluated for all positions that may occur during handling,
turning, and storage. Some techniques for preventing
damage to precast concrete units are:
a. additional conventional reinforcement.
b. external steel beams bolted to the unit.
c. handling units in pairs.
d. temporary additional prestress.
e. cables on each side of the member, along with queen
posts.
12.5.4 Design for erection loads—Particularly for
segmental construction, the designer should consider the
erection sequence and equipment loads which could be
applied to the structure. The erection sequence directly
affects the structural deflections. The magnitude of this
effect depends on the age of concrete when loaded and the
sequence of applying prestress loads.
Equipment reactions and segment weights can be substan-
tial during construction. In many instances where construc-
tion is done from above, e.g., erection with a launching
gantry or delivering segments over completed portions of the
structure, the stresses during erection may be the largest to
which the structure will ever be subjected.
12.5.5 Creep, shrinkage, and dead load deflection—Creep
of concrete is inversely related to the age of concrete at
loading. Consequently, deflection due to creep can be
reduced when loads are applied to older concrete. In this
sense, precast concrete members have an advantage over
cast-in-place construction because members can be precast
well in advance of load application.
Deflections due to creep and shrinkage may adversely
affect the serviceability of a bridge. For continuous struc-
tures, such deflections may produce undesirable secondary
stresses. Differential deflections of adjacent members may
cause unexpected overstress.
In order to predict creep and shrinkage, many variables
have to be known or assumed, as covered in section 5.4.2.
Simple formulas for determining creep and shrinkage are
provided by the AASHTO bridge specifications (Reference
HB-15, section 9.16.2.1); these are intended for highway
bridges (see also Reference 12-15).
12.5.6 Crown and superelevation—Whenever possible,
bridge seats are level, and the precast concrete units are set
vertically. For units which are relatively wide at the top, the
“fillets” may contain a considerable volume of cast-in-place
concrete, as illustrated in Fig. 12.5.6.1. Fillets are the spaces
between the bottom of cast-in-place deck and the horizontal
top surfaces of the precast units. Since the camber of
prestressed beams is not constant, the thickness of fillets
varies from one beam to the next. Field adjustment of fillet
thickness is done routinely. For large superelevation (more
than 5 percent) combined with wide units, such as standard
bulb tees or spread boxes, extra formwork and concrete for
the fillets may entail unwanted expense. Obviously, for high
superelevation, shorter spans with shapes having smaller top
width can be advantageous.
Except for narrow roadways, butted box beams are
installed perpendicular to the crown or superelevation, as
shown in Fig. 12.5.6.2. They have to be anchored securely to
avoid lateral creep.
12.6—Construction
12.6.1 Manufacturing—Precast concrete members can be
manufactured by firms regularly engaged in the production
of precast concrete in existing plants or at a specially
constructed job site plant. Existing precasting facilities will
usually provide superior facilities, trained workers, and
established quality control procedures for the materials and
manufacturing operations. A job site precasting plant may
prove necessary when extremely large precast segments are
required, or where transportation costs are excessive. Job site
precasting can result in significant cost savings for large
construction projects.
12.6.2 Transportation and erection—The ability to trans-
port and erect often determines the size and shape of the
 BRIDGE ANALYSIS AND DESIGN
Fig. 12.5.6.2—Superelevated deck with butted box beams.
precast elements (see Reference 12-6, pp. 734-750). Units of
moderate weight and size can be transported economically by
trucks and erected by crane. Rail transportation, supplemented
by other means, may be used for long distance shipments and
heavy or over-sized segments. Barge transportation may be the
most economical and practical for movement to water sites.
Erection considerations involve the availability of cranes,
derricks, launching gantries, site restrictions, and others.
12.6.3 Joints and connections—Properly detailed and
constructed joints and connections are essential to the
success of precast concrete bridge construction. Joints and
connections should be designed to transmit all forces and,
furthermore, be feasible to construct under actual job site
conditions. Since visible joints affect the appearance of the
bridge structure, well designed joints will enhance the struc-
ture’s esthetics.
12.6.4 Falsework—Falsework should be designed for
both primary and secondary effects with an adequate margin
of safety. For instance, when continuous structures are post-
tensioned, the post-tensioning forces may induce secondary
moments and change the magnitude of reactions. Falsework
should be capable of supporting any increase in applied
loads due to secondary effects.
Care should be taken in the design of the foundation of
falsework in order to avoid both excessive and uneven settle-
ment. To compensate for settlement, provisions for vertical
adjustment should be provided.
Temporary bents often have a significant effect on the
behavior of a bridge. When a temporary bent is removed, there
is a redistribution of stresses which should be evaluated and,
if significant, these stresses must be considered in the design.
RECOMMENDED REFERENCES
The documents of the various standards-producing organi-
zations referred to in this report are listed below with their
serial designation.
American Railway Engineering Association
Manual for Railway Engineering
American Association of State Highway and Transportation
Officials
HB15 Standard Specifications for Highway Bridges
343R-147
GSCB Guide Specifications for Design and Construc-
tion of Segmental Concrete Bridges
GSCBS Guide Specifications for Thermal Effects in
Concrete Bridge Superstructures
Precast/Prestressed Concrete Institute
STD 101 Standard Prestressed Concrete Beams for
Highway Bridge Spans 30 to 140 Ft
STD 107 Standard Prestressed Box Beams for Highway
Bridge Spans to 103 Ft
STD 108 Standard Prestressed Concrete Slabs for
Highway Bridge Spans to 55 Ft
STD 114 Prestressed Concrete Channel Slabs for Short
Span Bridges
STD 115 Standard Prestressed Concrete Bulb-Tee Beams
for Highway Bridge Spans to 150 ft
JR-343 Recommended Practice for Precast/Prestressed
Concrete Composite Bridge Deck Panels
CITED REFERENCES
12-1. “Precast Concrete Elements for Transportation
Facilities,” NCHRP Synthesis No. 53, Transportation
Research Board, Washington, D.C., 1978.
12-2. “Tentative Standards for Prestressed Concrete Piles,
Slabs, I-Beams, and Box Beams for Bridges and Interim
Manual for Inspection of Such Construction,” by Joint
Committee of AASHO and PCI; AASHO, 1963, 28 pp.
12-3. Heins, Conrad P., and Lawrie, Richard A., “Design
of Modern Concrete Highway Bridges,” John Wiley & Sons,
New York, 1984, pp. 362-470.
12-4. Lin, T.Y., and Burns, Ned H., “Design of Prestressed
Concrete Structures,” Third Edition, John Wiley & Sons,
New York, 1981.
12-5. Gerwick, Ben C., Jr., “Construction of Prestressed
Concrete,” Second Edition, Wiley Interscience, New York,
1993.
12-6. Libby, James R., “Modern Prestressed Concrete,”
Fourth Edition, Van Nostrand Reinhold Co., New York,
1990, pp. 620-640 and 734-750.
12-7 Oesterle, R. G.; Glikin, J. D.; and Larson, S. C.,
“Design of Precast Prestressed Bridge Girders Made Contin-
uous,” NCHRP Report 322, Transportation Research Board,
Washington, D.C., Nov. 1989.
343R-148 MANUAL OF CONCRETE PRACTICE
12-8. Freyermuth, C. L., “Design of Continuous Highway
Bridges with Precast, Prestressed Concrete Girders,” PCI
Journal, V. 14, No. 2, Apr. 1969.
12-9. Stanton, J. F., and Mattock, A. H., “Load Distribution
and Connection Design for Precast Stemmed Multibeam
Bridge Superstructures,” NCHRP Report 287, Transportation
Research Board, Washington, D.C., Nov. 1986.
12-10. PCI Design Handbook, Third Edition,
Precast/Prestressed Concrete Institute, Chicago, 1985.
12-11. “Fargo-Moorhead Toll Bridge,” PCI Journal, V. 14,
No. 3, May-June 1989.
12-12. “A Sound Investment: Trestle at $23 per Square
Foot,” Engineering News-Record, V. 221, No. 8, Aug. 25,
1988, p. 30.
12-13. Ingerslev, L. C. F., “Precast Concrete for the
Bahrain Causeway,” Concrete International, V. 11, No. 12,
Dec. 1989.
12-14. Abdel-Karim; Ahmad M.; and Tadros, Maher, K.
“Stretched-Out Precast I-Girder Bridge Spans,” Concrete
International, V. 13, No. 9, Sept. 1991.
12-15. “Prediction of Creep, Shrinkage, and Tempera-
ture Effects in Concrete Structures (ACI 209R-82(86)),”
American Concrete Institute, Farmington Hills, Mich.,
1986.
 BRIDGE ANALYSIS AND DESIGN
Reinforcing bars in box girder bridge with post-tensioning cables in girders.
CHAPTER 13—DETAILS OF REINFORCEMENT FOR DESIGN AND CONSTRUCTION
13.1—General
Details of reinforcement and bar supports not covered in
this chapter should be in accordance with ACI SP-66 or the
“Manual of Standard Practice,13-1” published by the
Concrete Reinforcing Steel Institute.
13.2—Development and splices of reinforcement
13.2.1 Development of reinforcement—General—The
calculated tension or compression in the reinforcement at
each section should be developed on each side of that section
by embedment length, hooks or mechanical devices, or a
combination thereof. Hooks may be used in developing rein-
forcing bars in tension only.
Tension reinforcement in flexural members may be devel-
oped by bending it across the web, or by making it contin-
uous with the reinforcement on the opposite face of the
member, or by anchoring it there.
The critical sections for development of reinforcement in
flexural members are at points of maximum stress and at
points within the span where adjacent reinforcement termi-
nates, or is bent. The recommendations in Section 13.2
should be followed.
Reinforcement should extend beyond the point at which it
is no longer required to resist flexure for a distance equal to
the effective depth of the member or 12db (12 nominal bar
diameters), whichever is greater, except at supports of
simple spans and at the free end of cantilevers.
Continuing reinforcement should have an embedment
length not less than the development length ld beyond the
point where bent or terminated tension reinforcement is no
longer required to resist flexure.
Flexural reinforcement should not be terminated in a tension
zone unless one of the following conditions is satisfied:
343R-149
a. The shear at the cutoff point does not exceed two-thirds
of that permitted, including the shear strength of the
web reinforcement provided.
b. The stirrup area in excess of that required for shear and
torsion is provided along each terminated bar or wire
over a distance from the termination point equal to
three-fourths the overall depth of the member. The
excess stirrup area Av should not be less than 60bw s/fy
or (0.4bw s/fy), where bw is the web width of the flexural
member, s is the stirrup spacing, and fy is the specified
yield strength. The spacing s should not exceed d/8βb,
where βb is the ratio of the area of the reinforcement cut
off to the total area of tension reinforcement at the
section.
For #11 (#35) and smaller bars, the continuing bars
provide double the area required for flexure at the cutoff
point, and the shear does not exceed three-fourths of that
permitted.
13.2.2 Development of positive moment reinforcement—
At least one-third of the positive moment reinforcement in
simple members and one-fourth of the positive moment rein-
forcement in continuous members should extend along the
same face of the member into the support. In beams, such
reinforcement should extend into the support at least 6 in.
(150 mm).
When a flexural member is part of the lateral load resisting
system, the positive moment reinforcement required to be
extended into the support should be anchored to develop the
specified yield strength fy in tension at the face of the support.
At simple supports and at points of inflection, positive
moment tension reinforcement should be limited to a diam-
eter such that ld computed for fy by Section 13.2 satisfies
343R-150 MANUAL OF CONCRETE PRACTICE
Eq. (13-1), except Eq. (13-1) need not be satisfied for rein-
forcement terminating beyond the centerline of simple
supports by a standard hook, or a mechanical anchorage
equivalent to a standard hook
ld ≤ Mn /Vu + la (13-1)
where
Mn = nominal moment strength assuming all reinforce-
ment at the section to be stressed to the specified
yield strength fy
Vu = factored shear force at the section
la = at a support, embedment length beyond the center
of the support
la = at point of inflection, limited to the effective depth
of the member or 12 db, whichever is greater
The value at Mn /Vu may be increased 30 percent when the
ends of the reinforcement are confined by a compressive
reaction.
13.2.3 Development of negative moment reinforcement—
Negative moment reinforcement in a continuous, restrained,
or cantilever member, or in any member of a rigid frame,
should be anchored in or through the supporting member by
embedment length, hooks, or mechanical anchorage. Nega-
tive moment reinforcement should have an embedment
length into the span as required by Section 13.2.
At least one-third of the total tension reinforcement
provided for negative moment at a support should have an
embedment length beyond the point of inflection, not less
than the effective depth of the member 12db or one-sixteenth
of the clear span, whichever is greater. This recommendation
accounts for a possible shifting of the moment diagram due
to the changes in loading, settlement of supports, or other
causes, in addition to the typically approximate nature of
moment calculations. If these calculations are carried out to a
higher degree of accuracy and accounted for possible moment
shifting, this distance could be justifiably decreased.
13.2.4 Development of reinforcement in special
members—Adequate anchorage should be provided for
tension reinforcement in flexural members where reinforce-
ment stress is not directly proportional to factored moment,
such as sloped, stepped, or tapered footings; brackets; deep
flexural members; or members in which the tension rein-
forcement is not parallel to the compression face.
13.2.5 Development length of deformed bars and deformed
wire in tension—The development length ld in in. (mm) of
deformed bars and deformed wire in tension should be
computed as the product of the basic development length of (a)
and the applicable modification factor of factors of (b), (c), and
(d), but ld should not be less than recommended in (e):
a. The basic development length should be:
For #11 (#35) and smaller bars (Note 1):
0.04A b f y ⁄ f c ′ or ( 0.02A b f y ⁄ f c ′ )
but not less than (Note 2):
0.0004 db fy or (0.06 dbfy)
For #14 (#45) bars (Note 3):
0.085f y ⁄ f c ′ or ( 0.025f y ⁄ f c ′ )
For #18 (#55) bars (Note 3):
0.11f y ⁄ f c ′ or ( 0.35f y ⁄ f c ′ )
For deformed wire:
0.03d b f y ⁄ f c ′ or [ ( 3d b f y ⁄ 8 ) f c ′ ]
where fc′ is the specified compressive strength of the
concrete, and Ab is the area of an individual bar.
b. The basic development length should be multiplied by
the applicable factor or factors for:
Top reinforcement (Note 4)................................ 1.4
Reinforcement with fy greater than 60,000 psi (400
MPa).............................. 2 - 60,000/fy or (2 – 400/fy).
c. Lightweight aggregate concrete:
When fct (average splitting tensile strength of light-
weight aggregate concrete) is specified and concrete is
proportioned in accordance with Section 3.2, the basic
development length may be multiplied by 6.7fc′/fct or
(fc′/1.8fct), but not less than 1.0.
When fct is not specified, the basic development length
should be multiplied by 1.33 for “all lightweight”
concrete or by 1.18 for “sand-lightweight” concrete.
Linear interpolation may be applied when partial sand
replacement is used.
d. A recent research project investigating the anchorage
of epoxy-coated reinforcing bars shows that the bond
strength is reduced when the coating prevents adhesion
between the bar and the concrete.13-2 At the time this
report was being prepared, ACI Committee 318 was in
the process of evaluating proposed changes to the ACI
318 Building Code, which included, as a result of this
research project, the addition of tension development
length modification factors for epoxy-coated rein-
forcing bars. According to results of this research
project, the modification factor for the tension develop-
ment length of epoxy-coated reinforcing bars should be
1.5 if the bars have either a clear cover of less than 3 bar
diameters or the clear spacing between the bars is less than
6 bar diameters, otherwise the modification factor should
be 1.2. The report also noted that the product of the factor
for “top reinforcement” and the factor for “epoxy-coated
reinforcement” need not be greater than 1.7.
e. The basic development length may be multiplied by the
applicable factor(s) for:
Reinforcement being developed in the length under
consideration and spaced laterally at least 6 in. (150
mm) on center and at least 3 in. (70 mm) clear from
the face of the member to the edge bar, measured in
the direction of the spacing ................................ 0.8.
Reinforcement in a flexural member in excess of
that required by analysis, where anchorage or devel-
opment for fy is not specifically required
...................................... (As required)/(As provided).
Reinforcement enclosed within a spiral not less than
1/4-in.- (5-mm-) diameter and not more than 4-in.-
(100-mm-) pitch .............................................. 0.75.
f. The development length ld should not be less than 12
in. (300 mm), except in the computation of lap splices
 BRIDGE ANALYSIS AND DESIGN 343R-151

by Section 13.2.15 and development of web reinforce-

ment by Section 13.2.
Notes:
1. The constant carries the unit of 1/in. (1/mm).
2. The constant carries the unit of in.2/lb (mm2/N).
3. The constant carries the unit of in. (mm).
4. Top reinforcement is horizontal reinforcement placed so
that more than 12 in. (300 mm) of concrete is cast in the
member below the reinforcement.
13.2.6 Development length of deformed bars in compres-
sion—The basic development length lbd for deformed bars in
compression should be computed as 0.02 db fy / f c ′ or
(dbfy /4) f c ′ , but should be not less than 0.0003dbfy or
(0.04db fy) when the constant 0.0003 (0.04) carries the unit of
in.2/lb (mm2/N). Where bar area in excess of that required by
analysis is provided, the basic development length may be
multiplied by the ratio of area required to area provided. The
basic development length may be reduced by 25 percent
when the bars are enclosed by spirals not less than 1/4-in.-
(5-mm-) in diameter and not more than 4-in.-(100-mm-)
pitch. The final development length should not be less than
8 in. (200 mm). Hooks should not be considered effective in
developing bars in compression.
13.2.7 Development length of bundled bars—The devel-
opment length of individual bars within a bundle, in tension,
or compression, should be that for the individual bar,
increased by 20 percent for a three-bar bundle, and 33 Fig. 13.2.8—Hooked bar details.
percent for a four-bar bundle.
13.2.8 Development of standard hooks in tension—The
than 3db, where db is the diameter of the hooked bar
development length ldh in in. (mm) for deformed bars in
........................................................................... 1.0.
tension terminating in a standard hook (Section 13.6) should
Where reinforcement is in excess of that required by
be computed as the product of the basic development length
analysis and anchorage or development for fy is not
lhb and the applicable modification factor or factors, but ldh
specifically required .... (As required)/(As provided).
should not be less than 8db or 6 in. (150 mm), whichever is
Lightweight aggregate concrete ........................ 1.3.
greater.
For details of hooks, ties, or stirrups and the measurement
The basic development length lhb for a hooked bar with fy of ldh, see Fig. 13.2.8.
equal to 60,000 psi (400 MPa) should be 1200db/ f c ′ or 13.2.9 Development length combination—Development
(100db/ f c ′ ), where the constant 1200 (100) carries the unit of reinforcement may consist of a combination of a hook or
of lb/in.2 (N/mm2). mechanical anchorage plus additional embedment length of
The basic development length lhb should be multiplied by reinforcement between the point of maximum bar stress and
the applicable factor(s): the mechanical anchorage.
For all bars with fy other than 60,000 psi (400 13.2.10 Development of welded wire fabric—The yield
MPa)....................................... fy /60,000 or (f /400). strength of welded smooth wire fabric may be considered
For #11 (#35) bars and smaller, side cover (normal developed by embedment of two cross wires with the closer
to plane of the hook) not less than 2-1/2 in. (60 mm), cross wire not less than 2 in. (50 mm) from the point of crit-
and for 90 deg hooks with cover on the bar extension ical section. However, the development length ld measured
beyond the hook not less than 2 in. (50 mm) ...... 0.7. from the point of critical section to the outermost cross wire
For #11 (#35) bars and smaller, with hook enclosed should not be less than 0.27Aw fy /sw f c ′ or (3.3Aw fy/sw f c ′ ),
vertically or horizontally within ties or stirrup ties where Aw is the area of an individual wire, and sw is the
spaced not greater than 3db, where db is the diameter spacing of the wires to be developed. The development
of the hooked bar ............................................... 0.8. length ld may be modified by (As required)/(As provided) for
For bars being developed by a standard hook at reinforcement in excess of that required by analysis, and by
discontinuous ends of members with both side cover the factor of Section 13.2 for lightweight aggregate concrete,
and top (or bottom) cover over the hook less than 2-1/2 but should not be less than 6 in. (150 mm), except in the
in. (60 mm), and for a 90 deg hook only, end cover computation of lap splices by Section 13.2.
less than 2 in. (50 mm), the hooked bar should be The development length ld in in. (mm), of welded
enclosed within ties or stirrups spaced not greater deformed wire fabric measured from the point of critical
343R-152 MANUAL OF CONCRETE PRACTICE
section to the end of the wire should be computed as the
product of the basic development length of this section and
the applicable modification factor(s) of Section 13.2, but ld
should not be less than 8 in. (200 mm), except in the compu-
tation of lap splices by Section 13.2 and development of web
reinforcement by Section 13.2.
The basic development length of welded deformed wire
fabric, with at least one cross wire within the development
length not less than 2 in. (50 mm) from the point of critical
section should be
0.03d b ( f y – 20, 000 ) ⁄ f c ′ or 3d b ( f y – 140 ⁄ 8 ) ⁄ ( f c ′ )
but not less than
0.20A w f y ⁄ ( s w f c ′ ) or 2.5A w f y ⁄ ( s w f c ′ )
where the constants 20,000 and 140 have units of psi and
MPa, respectively.
The basic development length of welded deformed wire
fabric, without cross wires within the development length
should be determined the same as deformed wire.
13.2.11 Development length of prestressing strand—
Three- or seven-wire pretensioning strand should be bonded
beyond the critical section for a development length in in.
(mm), not less than
(fps – 2fse /3)db or [(fps – 2fse /3)db/7]
where db is the strand diameter in in. (mm) fps (stress in pre-
stressed reinforcement at nominal strength) and fse (effective
stress in prestressed reinforcement) are expressed in kips per
square in. (MPa); and the expression in the parenthesis is
used as a constant without units. Investigation may be limit-
ed to cross sections nearest each end of the member that are
required to develop full design strength under the specified
factored loads.
Where bonding of a strand does not extend to the end of a
member, and the design includes tension at service loads in
the precompressed tensile zone as permitted by Section 9.5,
the development length previously recommended should be
doubled.
13.2.12 Mechanical anchorage—Any mechanical device
capable of developing the strength of the reinforcement
without damage to the concrete may be used as anchorage.
13.2.13 Development of web reinforcement—Web rein-
forcement should be carried as close to the compression and
tension surfaces of a member as cover requirements and the
proximity of other reinforcements permit. The ends of single
leg, single U-, or multiple U-stirrups should be anchored by
one of the following means:
a. A standard hook plus an embedment of 0.5ld. The 0.5ld
embedment of a stirrup leg should be the distance
between the middepth of the member h/2 and the start
of the hook (point of tangency).
b. An embedment of h/2 above or below the middepth on
the compression side of the member for a full develop-
ment length ld, but not less than 24db, or 12 in. (300
mm) for deformed bars or wire.
c. For #5 (#15) bars and D31 wire and smaller, bending
around the longitudinal reinforcement through at least
135 deg (2.35 rad.), plus for stirrups with design stress
exceeding 40,000 psi (300 MPa), an embedment of
0.33ld. The 0.33ld embedment of a stirrup leg should be
the distance between the middepth of the member h/2
and the start of the hook (point of tangency).
d. For each leg of welded smooth wire fabric forming
single U-stirrups, either two longitudinal wires spaced
2 in. (50 mm) along the member at the top of the U, or
one longitudinal wire located not more than d/4 from
the compression face, and a second wire closer to the
compression face and spaced not less than 2 in. (50
mm) from the first wire. The second wire may stirrup
leg beyond a bend, or on a bend which has an inside
diameter of bend not less than 8db.
e. For each end of a single leg stirrup of welded smooth
or deformed wire fabric, two longitudinal wires at a
maximum spacing of 2 in. (50 mm) and the inner wire
at least the greater of d/4 or 2 in. (50 mm) from the
middepth of the member. The outer longitudinal wire at
the tension face should not be farther from the face than
the portion of primary flexural reinforcement closest to
the face.
f. Pairs of U-stirrups or ties placed to form a closed unit
should be considered properly spliced when the lengths
of laps are 1.7ld. In members at least 18-in.- (500-mm-
) deep, splices with Abfy not more than 9000 lb (40 kN)
per leg should be considered adequate if stirrup legs
extend the full available depth of the member.
Between the anchored ends, each bend in the continuous
portion of a single U- or multiple U-stirrup should enclose a
longitudinal bar. Longitudinal bars bent to act as shear rein-
forcement, if extended into a region of tension, should be
continuous with the longitudinal reinforcement, and if
extended into a region of compression, should be anchored
beyond the middepth of the member as specified for devel-
opment length in Section 13.2 for that part of fy required to
satisfy Eq. (7-60).
13.2.14 Splices of reinforcement—General—Splices of
reinforcement should be made only as required or permitted
on the design drawings, or in the specifications, or as autho-
rized by the engineer.
Lap splices should not be used for bars larger than #11
(#35). Bars #14 and #18 (#45 and #55) may be lap-spliced in
compression only to #11 (#35) and smaller bars, and in
compression only to smaller size footing dowels.
Lap splices of bundled bars should be based on the lap
splice length recommended for individual bars within a
bundle, and such individual splices within the bundle should
not overlap. The length of lap for bundled bars should be
increased 20 percent for a three-bar bundle and 33 percent
for a four-bar bundle. Bars spliced by noncontact lap splices
in flexural members should not be spaced transversely
farther apart than one-fifth the required lap splice length, or
6 in. (150 mm).
Welded splices and mechanical connections may be used
unless otherwise prohibited or restricted. A full-welded
 BRIDGE ANALYSIS AND DESIGN
splice is one in which the bars are butted and welded to
develop in tension at least 125 percent of the specified yield
strength fy of the bar. A full mechanical connection is one in
which the bars are connected to develop in tension or
compression at least 125 percent of the specified yield
strength fy of the bar. Welded splices and mechanical
connections not meeting these recommendations may be
used in accordance with Section 13.2.
When required or permitted, all welding of reinforcing
bars should conform to AWS D1.4, “Structural Welding
Code—Reinforcing Steel.”13-3 Welding of wire to wire, and
welding of wire or welded wire fabric to reinforcing bars or
structural steels, should conform to applicable provisions of
AWS D1.4 and supplementary requirements specified by the
engineer. The engineer should also specify any desired or
stringent requirements for preparation or welding, such as
the removal of zinc or epoxy coating than those contained in
AWS D1.4; any desired or stringent requirements for chem-
ical composition of reinforcing bars than those contained in
the referenced ASTM Specifications; and special heat treat-
ment of welded assemblies if required.
13.2.15 Splices of deformed bars and deformed wire in
tension—The minimum length of lap for tension lap splices
should be as required for Class A, B, or C splices, but not less
than 12 in. (300mm), where
Class A splice .................................................... 1.0ld
Class B splice .................................................... 1.3ld
Class C splice .................................................... 1.7ld
where ld is the tension development length for the specified
yield strength fy calculated in accordance with Section 13.2.
Lap splices of deformed bars and deformed wire in tension
should conform to Table 13.2.15.
Welded splices or mechanical connections used where the
area of reinforcement provided is less than twice that required
by analysis should meet the requirements of Section 13.2.
Welded splices or mechanical connections used where the
area of reinforcement provided is at least twice that required
by analysis should meet the following:
a. Splices should be staggered at least 24 in. (600 mm)
and in a manner to develop at every section at least
twice the calculated tensile force at that section, but not
less than 20,000 psi (140 MPa) for the total area of rein-
forcement provided.
b. In computing the tensile force developed at each
section, spliced reinforcement may be rated at the spec-
ified splice strength. Unspliced reinforcement should
be rated at that fraction of fy defined by the ratio of the
shorter actual development length to ld required to
develop the specified yield strength fy.
c. Splices in “tension tie members” should be made with
a full-welded splice or full mechanical connection in
accordance with Section 13.2, and splices in adjacent
bars should be staggered at least 30 in. (800 mm).
13.2.16 Splices of deformed bars in compression—The
minimum length of lap for compression lap splices for
deformed bars in compression should be the development
length in compression, computed in accordance with Section
13.2, but not less than
343R-153
Table 13.2.15—Tension lap splices
A s provided* Maximum percent of As spliced within required
--------------------------------
A s required lap length
50 75 100
Equal to or greater Class A Class A Class B
than 2
Less than 2 Class B Class C Class C
*Ration of area or reinforcement provided to area of reinforcement required by
analysis at the splice location.
For fy of 60,000 psi (400 MPa) or less:
0.0005fy db or (0.07fydb)
For fy greater than 60,000 psi (400 MPa):
0.0009fy – 24)db or [(0.13fy – 24)db]
or 12 in. (300 mm).
When the specified compressive strength of the concrete is
less than 3000 psi (20 MPa), the length of lap should be
increased by one-third.
When bars of different sizes are lap-spliced in compres-
sion, the splice length should be the larger of the compres-
sion development length of the larger bar or the lap splice
length of the smaller bar.
In tied compression members where ties throughout the
lap length have an effective area of at least 0.0015hs, where
h is the total depth of the member and s is the spacing of the
tie, the lap splice length may be 0.83 of that previously
recommended, but not less than 12 in. (300 mm). Tie legs
perpendicular to dimension h should be used in determining
the effective area.
For splices within the spiral of a spirally-reinforced
compression member, 0.75 of the lap splice length previ-
ously recommended may be used, but the lap length should
not be less than 12 in. (300 mm).
For end bearing splices in which the bars are required for
compression only, the compressive stress may be transmitted
by bearing of square cut ends held in concentric contact by a
suitable device. The ends of the bars should terminate in flat
surfaces within 1-1/2 deg of a right angle to the axis of the
bars and should be fitted within 3 deg of full bearing after
assembly. End bearing splices should be used only in
members having closed ties, closed stirrups, or spirals.
Welded splices or mechanical connections used in
compression should meet the recommendations for full
welded or full mechanical connections (Section 13.2).
13.2.17 Splices of welded deformed wire fabric in
tension—The minimum length of lap for lap splices of
welded deformed wire fabric measured between the ends of
each fabric sheet should not be less than 1.7ld or 8 in. (200 mm),
and the overlap measured between the outermost cross wires
of each fabric sheet should not be less than 2 in. (50 mm). ld
is the development length for the specified yield strength fy
computed in accordance with Section 13.2.
Lap splices of welded deformed wire fabric, without cross
wires within the lap splice length, should be determined the
same as for deformed wire.
343R-154 MANUAL OF CONCRETE PRACTICE
13.2.18 Splices of welded smooth wire fabric in tension—
The minimum length of lap for lap splices of welded smooth
wire fabric should be in accordance with the following:
a. When the area of reinforcement provided is less than
twice that required by analysis at the splice location,
the length of the overlap measured between the outer-
most cross wires of each fabric sheet should not be less
than one spacing of the cross wires plus 2 in. (50 mm),
or less than 1.5ld, or 6 in. (150 mm). ld is the develop-
ment length for the specified yield strength fy computed
in accordance with Section 13.2.
b. When the area of reinforcement provided is at least
twice that required by analysis at the splice location,
the length of the overlap measured between the outer-
most cross wires of each fabric sheet should not be less
than 1.5ld, or 2 in. (50 mm).
13.3—Lateral reinforcement for compression
members
13.3.1 Spirals—Spiral reinforcement for compression
members should consist of evenly spaced continuous spirals
held firmly in place by attachment to the longitudinal rein-
forcement, and held true to line by vertical spacers. The
spirals should be of size and assembled to permit handling
and placing without being distorted from the designed
dimensions. Spiral reinforcement may be plain or deformed
bars, smooth or deformed wire, with a minimum diameter of
3
/8 in. (10 mm). Anchorage of spiral reinforcement should be
provided by one and one-half extra turns of spiral bar or wire
at each end of the spiral unit. Splices in spiral reinforcement
should be lap splices of 48db, but not less than 18 in. (450
mm), or welded. The clear spacing between spirals should
not exceed 3 in. (80 mm), or be less than 1 in. (25 mm) or
one-and-one-third times the maximum size of coarse aggre-
gate used. Spirals should extend from the top of the footing
or other support to the level of the lowest horizontal rein-
forcement in the members previously supported.
The ratio of spiral reinforcement ρs should not be less than
the value given by ρs = 0.45 (Ag/Ac – 1)fc′/fy, where Ag is the
gross area of the section, Ac is the core area of the section
measured to the outside diameter of the spiral, and fy is the
specified yield strength of the spiral reinforcement not more
than 60,000 psi (400 MPa).
13.3.2 Ties—All nonprestressed bars for tied compression
members should be enclosed by lateral ties of the following
minimum sizes:
U.S. Customary:
#3 size ties (#10 longitudinal bars and smaller)
#4 size ties (#11, #14, #18 longitudinal bars or bundled
longitudinal bars)
Metric:
#10 size ties
Deformed wire or welded wire fabric of equivalent area
may be used. The vertical spacing of the ties should not
exceed 16 longitudinal bar diameters, 48 tie bar or wire
diameters, or the least dimension of the member. This
spacing may be increased in compression members which
have a larger cross section than required by conditions of
loading. When bars larger than #10 (#30) are bundled with
more than two bars in any one bundle, the tie spacing should
be reduced to one-half of that previously recommended. Ties
should be located vertically not more than half a tie spacing
above the top of the footing or other support and should be
spaced not more than half a tie spacing below the lowest
horizontal reinforcement in the members previously
supported. The ties should be arranged so that every corner
and alternate longitudinal bar has lateral support provided by
the corner of a tie having an included angle of not more than
135 deg, and no bar should be farther than 6 in. (150 mm) on
either side from a laterally supported bar. Where the bars are
located around the periphery of a circle, a complete circular
tie may be used.
13.3.3 Prestressing steel—Except when used in walls, all
prestressing steel should be enclosed by spirals as recom-
mended in Section 13.3 or by lateral ties at least #3 (#10) in
size and spaced as recommended in Section 13.3.
13.3.4 Oversized members—In a compression member
which has a larger cross section than required by conditions
of loading, the lateral reinforcement requirements may be
waived where structural analysis or tests show adequate
strength and feasibility of construction.
13.3.5 Seismic areas—In seismic areas, where an earth-
quake which could cause major damage to construction has
a high probability of occurrence, lateral reinforcement for
compression members should be designed and detailed to
provide adequate strength and ductility to resist expected
seismic movements.13-4,ACI 318
13.4—Lateral reinforcement for flexural members
13.4.1 Compression reinforcement—Compression rein-
forcement in beams should be enclosed by ties or stirrups,
satisfying the size and spacing recommendations in Section
13.3, or by welded wire fabric of an equivalent area. Such
ties or stirrups should be provided throughout the distance
where compression reinforcement is required.
13.4.2 Torsion or stress reversal—Lateral reinforcement
for flexural members, subject to stress reversals or to torsion
at supports, should consist of closed ties, closed stirrups, or
spirals extending around the flexural reinforcement.
Closed ties or stirrups may be formed in one piece by over-
lapping standard stirrup or tie end hooks around a longitu-
dinal bar, or formed in one or in two pieces, lap-spliced with
a Class C splice (lap of 1.7ld), or anchored in accordance
with the recommendations in Section 13.2.
13.4.3 Seismic areas—In seismic areas, where an earth-
quake that could cause major damage to construction has a
high probability of occurrence, lateral reinforcement should
be designed and detailed to provide adequate strength and
ductility to resist expected seismic movements.13-4
13.5—Shrinkage and temperature reinforcement
Reinforcement for shrinkage and temperature stresses
should be provided near exposed surfaces of walls and slabs
not otherwise reinforced. For Grade 60 (400 MPa) deformed
bars or welded wire fabric (smooth or deformed), a
minimum area of reinforcement equal to 0.18 percent of the
 BRIDGE ANALYSIS AND DESIGN
gross concrete area should be provided. The spacing of
stress-relieving joints should be considered in determining
the area of shrinkage and temperature reinforcement. The
preceding minimum area of reinforcement should be
increased proportionately for large joint spacings.
For Grade 40 (275 MPa) or 50 (345 MPa) deformed bars,
a minimum area of reinforcement equal to 0.20 percent of the
gross concrete area should be provided.
Shrinkage and temperature reinforcement should not be
spaced farther apart than three times the wall or slab thick-
ness, or 18 in. (500 mm).
At all required sections, reinforcement for shrinkage and
temperature stresses should develop the specified yield
strength fy in tension in accordance with the recommenda-
tions in Section 13.2.
Bonded or unbonded prestressing tendons may be used for
shrinkage and temperature reinforcement in structural slabs.
The tendons should provide a minimum average compres-
sive stress of 100 psi (0.7 MPa) on the gross concrete area,
based on effective prestress after losses. Spacing of tendons
should not exceed 6 ft (2 m). When the spacing is greater
than 54 in. (1.4 m), additional bonded reinforcement should
be provided.
Mass concrete is defined as any large volume of cast-in-
place concrete with dimensions large enough to require that
measures be taken to cope with the generation of heat and
attendant volume change to minimize cracking. For mass
concrete walls, slabs and footings less than 48-in.- (1.2-m-)
thick, minimum shrinkage and temperature steel should be
0.0015Ag. No less than one-half or more than two-thirds of
this total quantity should be placed in any one face.
Maximum bar spacing should be limited to 12 in. (300 m).
For members more than 48-in.- (1.2-m-) thick, minimum
shrinkage and temperature steel in each face should be such
that Ab = 2dcs/100, where Ab is the area of bar, dc is the
distance from the centroid of reinforcement to the concrete
surface, and s is the spacing of the bar (ACI 207.2R). The
minimum bar size and maximum spacing should be #6 (#20)
at 12 in. (65 mm) on center.
13.6—Standard hooks and minimum bend
diameters
13.6.1 Standard hooks—Standard hooks for end
anchorage of reinforcing bars should be either a 180 deg
bend plus a 4db extension, but not less than 2-1/2 in. (60 mm)
at the free end of the bar, or a 90 deg bend plus a 12db exten-
sion at the free end of the bar, where db is the nominal diam-
eter of the bar (see Fig. 13.2.8).
For stirrup and tie hooks only:
a. For #5 (#15) bars and smaller, a 90 deg bend plus a 6db
extension at the free end of the bar, or
b. For #6, #7 and #8 (#20 and #25) bars, a 90 deg bend
plus a 12db extension at the free end of the bar, or
c. For #8 (#25) bars and smaller, a 135 deg bend plus a
6db extension at the free end of the bar.
13.6.2 Minimum bend diameters—Main reinforcement—
The finished diameter of bend measured on the inside of the
bar, other than for stirrups and ties in sizes #3 through #5
343R-155
Table 13.6—Minimum diameters of bend
Bar size Minimum diameter
#3 through #8 (#10 through #25) 6 bar diameters
#9, #10, and #11 (#30 and #35) 8 bar diameters
#14 and #18 (#45 and #55) 10 bar diameters
(#10 and #15), should not be less than the values shown in
Table 13.6.
13.6.3 Minimum bend diameters—Ties and stirrups—The
finished inside diameter of bend for stirrups and ties should
not be less than four bar diameters for size #5 (#15) and
smaller, and six bar diameters for sizes #6 to #8 (#20 to #25),
inclusive. The inside diameter of bend in welded wire fabric,
smooth or deformed, for stirrups and ties should not be less
than four wire diameters for deformed wire larger than D6
and two wire diameters for all other wires. Bends with an
inside diameter of less than eight wire diameters should not
be less than four wire diameters from the nearest welded
intersection.
13.7—Spacing of reinforcement
13.7.1 Cast-in-place concrete—For cast-in-place
concrete, the clear distance between parallel bars and/or
tendon ducts in a layer should not be less than one and one-
half times the nominal diameter of the bars or tendon ducts,
one and one-half times the maximum size of the coarse
aggregate, or 1-1/2 in. (40 mm). When required or permitted
by the engineer, this minimum distance may be decreased.
13.7.2 Precast concrete—For precast concrete manufac-
tured under plant control conditions, the clear distance
between parallel bars in a layer should not be less than the
nominal diameter of the bars, one and one-third times the
maximum size of the coarse aggregate, or 1 in. (25 mm).
13.7.3 Multilayers—Where positive or negative reinforce-
ment is placed in two or more layers with the clear distance
between layers not more than 6 in. (150 mm), the bars in the
upper layers should be placed directly above those in the
bottom layer with the clear distance between the layers not
less than 1 in. (25 mm) or the nominal diameter of the bars.
13.7.4 Lap splices—The clear distance limitation between
bars should also apply to the clear distance between a contact
lap splice and adjacent splices or bars.
13.7.5 Bundled bars—Groups of parallel reinforcing bars
bundled in contact to act as a unit should be limited to four
in any one bundle. Bars larger than #11 (#35) should be
limited to two in any one bundle in flexural members;
bundled bars should be located within stirrups or ties. Indi-
vidual bars in a bundle cut off within the span of a flexural
member should terminate at different points with at least a
40db stagger. Where spacing limitations and minimum
concrete cover are based on bar size, a unit of bundled bars
should be treated as a single bar of a diameter derived from
the equivalent total area.
13.7.6 Walls and slabs—In walls and slabs, the principal
reinforcement should not be spaced farther apart than one
and one-half times the wall or slab thickness, or more than
18 in. (500 mm).
343R-156 MANUAL OF CONCRETE PRACTICE
13.7.7 Pretensioning steel—The clear distance between
pretensioning steel at the end of a member should not be less
than four times the diameter of individual wires, three times
the diameter of strands, or one and one-third times the
maximum size of the coarse aggregate. Closer vertical
spacing and bundling of strands may be permitted in the
middle portion of the span.
13.7.8 Post-tensioning ducts—The clear distance between
post-tensioning ducts at the end of a member should not be
less than 1-1/2 in. (40 mm), or one and one-third times the
maximum size of the coarse aggregate.
The inside diameter of the duct for a post-tensioning steel
bar, strand, or wire should be at least 1/4 in. (6 mm) greater
than the outer diameter of the post-tensioning steel. When
more than one bar, strand, or wire is used in a tendon, the
area of the duct should be at least two times the area of
prestressing steel within the duct.
Ducts for post-tensioning steel may be bundled if the
concrete can be satisfactorily placed and when provision is
made to prevent the steel, when tensioned, from breaking
through the duct. The clear distance limitation between ducts
previously recommended should be maintained in the end 3
ft (1 m) of a member.
13.8—Concrete protection for reinforcement
13.8.1 Minimum cover—The following minimum
concrete cover is recommended for prestressed and nonpre-
stressed reinforcement:
Minimum cover
in. (mm)
Concrete cast against and
permanently exposed to earth 3 (70)
Concrete exposed to earth or weather:
Principal reinforcement 2 (50)
Stirrups, ties, and spirals 1-1/2 (40)
Concrete bridge slabs:
Top reinforcement 2 (50)
Bottom reinforcement 1-1/2 (40)
Concrete not exposed to weather or
in contact with the ground:
Principal reinforcement 1-1/2 (40)
Stirrups, ties, and spirals 1 (30)
13.8.2 Bundled bars—For bundled bars, the minimum
concrete cover should be equal to the lesser of the equivalent
diameter of the bundle or 2 in. (50 mm), but not less than that
recommended in Section 13.8.
13.8.3 Corrosive environments—In corrosive or marine
environments or other severe exposure conditions, the
amount of concrete protection should be suitably increased,
and the denseness and nonporosity of the protecting concrete
should be considered. Coated reinforcement or other protec-
tion should be provided.
13.8.4 Future extensions—Exposed reinforcing bars,
inserts, and plates intended for bonding with future exten-
sions should be protected from corrosion.
13.9—Fabrication
Reinforcement should be fabricated to the shapes shown
on the design drawings. All bars should be bent cold unless
otherwise permitted by the engineer. Reinforcing bars
should be fabricated in accordance with the fabricating toler-
ances given in SP-66.
Placing drawings showing all fabrication dimensions and
locations for placing of reinforcement and bar supports
should be submitted for review and acceptance. Acceptance
should be obtained before fabrication.
Reinforcing bars should be shipped in bundles, tagged,
and marked in accordance with the recommendations in the
“Manual of Standard Practice.”13-1
13.10—Surface conditions of reinforcement
All reinforcement, at the time concrete is placed, should be
free of mud, oil, or other materials that may adversely affect
or reduce the bond. Research has shown that a normal
amount of rust increases bond. Normal rough handling
generally removes excessive rust, which would be loose
enough to reduce bond.
Reinforcement, except prestressing steel, with rust, mill
scale, or a combination of both should be considered satis-
factory, provided the minimum dimensions, including height
of deformations and weight of a hand-wire-brushed test
specimen are not less than the applicable ASTM specifica-
tion requirements (see Section 3.2.).
Prestressing steel should be clean and free of excessive
rust, oil, dirt, scale, and pitting. A light oxide is permissible.
13.11—Placing reinforcement
13.11.1 General—All reinforcement should be supported
and fastened before the concrete is placed, and should be
secured against displacement within the placing tolerances
permitted. The placing of bars should be in accordance with
the recommendations in “Placing Reinforcing Bars.”13-5
Reinforcing bars in the top layer of bridge decks should be
tied at all intersections, except where spacing is less than 1 ft
(300 mm) in each direction, then alternate intersections
should be tied.
Tolerances on placing reinforcement should conform to
ACI 117.
Substitution of different size or grade of reinforcing bars
should be permitted only when authorized by the engineer.
Reinforcing bars supported from formwork should rest on
bar supports made of concrete, metal, plastic, or other
acceptable materials. On ground or mud mat, supporting
concrete blocks may be used. Where the concrete surface
will be exposed to weather, the portions of all bar supports
within 1/2 in. (15 mm) of the concrete surface should be
noncorrosive or protected against corrosion.
13.11.2 Zinc-coated (galvanized) bars—Zinc-coated
(galvanized) reinforcing bars supported from formwork
should rest on galvanized wire bar supports, on wire bar
supports coated with dielectric material, or on bar supports
made of dielectric material or other acceptable materials. All
other reinforcement and embedded steel items in contact
with or in close proximity to galvanized reinforcing bars
 BRIDGE ANALYSIS AND DESIGN
should be galvanized, unless otherwise required or
permitted. They should also be fastened with zinc-coated or
nonmetallic-coated tie wire or other acceptable materials.
13.11.3 Epoxy-coated bars—Epoxy-coated reinforcing
bars supported from formwork should rest on coated wire
bar supports, on bar supports made of dielectric material, or
other acceptable materials. Wire bar supports should be
coated with dielectric material for a minimum distance of 2
in. (50 mm) from the point of contact with the epoxy-coated
reinforcing bars. Reinforcing bars used as support bars
should be epoxy-coated. In walls reinforced with epoxy-
coated bars, spread bars, where specified, should be epoxy-
coated. Proprietary combination bar clips and spreaders used
in walls with epoxy-coated reinforcing bars should be made
of corrosion-resistant material or coated with dielectric
material. Epoxy-coated reinforcing bars should be fastened
with epoxy-, plastic-, or nylon-coated tie wire or other
acceptable materials.
13.11.4 Welded wire fabric—Welded wire fabric rein-
forcement should be supported as recommended for rein-
forcing bars.
13.11.5 Splices—All reinforcement should be furnished in
full lengths as indicated on the design drawings, or in the spec-
ifications. Splices in reinforcement not indicated on the design
drawings or in the specifications, should be permitted only
when authorized by the engineer. All splices should be in accor-
dance with the recommendations given in Section 13.2.
13.11.6 Welding—When required or acceptable, all
welding of reinforcing bars should be in accordance with the
recommendations given in Section 13.2. Welding of crossing
bars (tack welding) should not be permitted for assembly or
reinforcement unless authorized by the engineer.
After completion of welding on zinc-coated (galvanized)
or epoxy-coated reinforcing bars, coating damage should be
repaired in accordance with the recommendations given in
Section 3.2. All welds and all steel splice members used to
splice bars should be coated with the same material used for
repair of coating damage. Suitable ventilation should be
provided when welding zinc-coated (galvanized) or epoxy-
coated reinforcing bars.
13.11.7 Mechanical connections—Mechanical connec-
tions should be installed in accordance with the manufac-
turer's recommendations, and as accepted by the engineer.
After installation of mechanical connections on zinc-coated
(galvanized) or epoxy-coated reinforcing bars, coating
damage should be repaired in accordance with the recom-
mendations given in Section 3.2. All parts of mechanical
connections used on coated bars, including steel splice
sleeves, bolts, and nuts should be coated with the same mate-
rial used for repair of coating damage.
13.11.8 Field bending and cutting—Reinforcing bars
partially embedded in concrete should not be field bent
except as indicated in the Contract Documents or when
permitted by the engineer. If zinc-coated (galvanized) or
epoxy-coated reinforcing bars are field bent, coating damage
should be repaired in accordance with the recommendations
given in Section 3.2.
343R-157
Coated reinforcing bars should not be cut in the field
except when permitted by the engineer. When zinc-coated
(galvanized) bars or epoxy-coated bars are cut in the field,
the ends of the bars should be coated with the same material
used for repair of coating damage.
13.11.9 Storage and handling of coated reinforcing
bars—Coating damage due to handling, shipment, and
placing should be repaired in accordance with the recom-
mendations given in Section 3.2.
Equipment for handling epoxy-coated reinforcing bars
should have protected contact areas. Bundles of coated bars
should be lifted at multiple pick-up points to prevent bar-to-
bar abrasion from sags in the bundles. Coated bars or bundles
of coated bars should not be dropped or dragged. Coated bars
should be stored on protective cribbing. Fading of the color
of the coating should not be cause for rejection of epoxy-
coated reinforcing bars. Coating damage due to handling,
shipment, and placing need not be repaired in cases where
the damaged area is 0.1 in.2 (60 mm2) or smaller. Damaged
areas larger than 0.1 in.2 (60 mm2) should be repaired in
accordance with Section 3.2. The maximum amount of
damage, including repaired and unrepaired areas, should not
exceed 2 percent of the surface area of each bar.
13.12—Special details for columns
13.12.1 Offsets—Where longitudinal bars are offset bent,
the slope of the inclined portion of the bar with the axis of the
column should not exceed 1 in 6, and the portions of the bar
above and below the offset should be parallel to the axis of
the column. Adequate horizontal support at the offset bends
should be treated as a matter of design, and should be
provided by lateral ties, spirals, or parts of the construction.
Lateral ties or spirals so designed should be placed not more
than 6 in. (150 mm) from points of bend. The horizontal
thrust to be resisted should be assumed as one and one-half
times the horizontal component of the computed force in the
inclined portion of the bar.
Offset bars should be bent before they are placed in the forms
(see Section 13.9). Bundled bars should not be offset bent.
Where a column face is offset 3 in. (70 mm) or more,
longitudinal bars should not be offset bent. Splices of longi-
tudinal bars adjacent to the offset column faces should be
made by separate dowels lapped as required herein.
13.12.2 Splices—Where the factored load stress in the
longitudinal bars in a column, calculated for various loading
conditions, varies from fy in compression to 1/2fy or less in
tension, lap splices, butt welded splices, mechanical connec-
tions, or end bearing splices may be used. The total tensile
strength provided in each face of the column by the splices
alone, or by the splices in combination with continuing
unspliced bars at the specified yield strength fy, should be at
least twice the calculated tension in that face of the column
but not less than required by Section 13.12.
Where the factored load stress in the longitudinal bars in a
column, calculated for any loading condition exceeds 1/2fy in
tension, lap splices designed to develop the specified yield
strength fy in tension, or full welded splices or full mechan-
ical connections should be used.
343R-158 MANUAL OF CONCRETE PRACTICE
At horizontal cross sections of columns where splices are
located, a minimum tensile strength at each face of the
column equal to one-fourth the area of vertical reinforcement
in that face multiplied by fy should be provided.
13.12.3 Composite columns—Structural steel cores in
composite columns should be accurately finished to bear at
end bearing splices, and positive provision should be made
for alignment of one core above the other in concentric
contact. Bearing may be considered effective to transfer 50
percent of the total compressive stress in the steel core. At
the column base, provision should be made to transfer the
load to the footing, in accordance with Section 11.3. The
base of the structural steel section should be designed to
transfer the total load from the entire composite column to
the footing, or it may be designed to transfer the load from
the steel section only, provided it is placed to leave an ample
section of concrete for transfer of the portion of the total load
from the reinforced concrete section of the column to the
footing by means of bond on the vertical reinforcement and
by direct compression on the concrete.
RECOMMENDED REFERENCES
The documents of the various standards-producing organi-
zations referred to in this report are listed below with their
serial designation, including year of adoption or revision.
The documents listed were the latest effort at the time this
report was written. Since some of these documents are
revised frequently, generally in minor detail only, the user of
this report should check directly with the sponsoring group
if it is desired to refer to the latest revision.
American Concrete Institute
SP-66 ACI Detailing Manual—1980
117-81 Standard Tolerances for Concrete
Construction and Materials
318-83 Building Code Requirements for
Reinforced Concrete
207.2R-73 Effect of Restraint, Volume Change and
(Reaffirmed 1980) Reinforcement on Cracking of Massive
Concrete
CITED REFERENCES
13-1. Manual of Standard Practice, CRSI Publication
MSP-1-86, Concrete Reinforcing Steel Institute, Schaum-
burg, IL, 1986.
13-2. Treece, R. A., and Jirsa, J. O., “Bond Strength of
Epoxy-Coated Reinforcing Bars,” Research Project spon-
sored by the Concrete Reinforcing Steel Institute and the
Reinforced Concrete Research Council, Department of Civil
Engineering, The University of Texas at Austin, 1987.
13-3. “Structural Welding Code—Reinforcing Steel,”
AWS Publication D1.4-79, American Welding Society,
Miami, 1979.
13-4. “Seismic Design Guidelines for Highway Bridges
(ATC-6),” Applied Technology Council, Berkeley, CA, Oct.
1981.
13-5. Placing Reinforcing Bars, Fifth Edition, Concrete
Reinforcing Steel Institute, Schaumburg, IL, 1986.
ACI 343R-95 was submitted to letter ballot of the committee and approved in accor-
dance with ACI Balloting procedures.